Next Article in Journal
Countering Negative Effects of Terrain Slope on Airborne Laser Scanner Data Using Procrustean Transformation and Histogram Matching
Next Article in Special Issue
Mixed-Severity Fire Fosters Heterogeneous Spatial Patterns of Conifer Regeneration in a Dry Conifer Forest
Previous Article in Journal
Effects of Climate Change on the Potentially Suitable Climatic Geographical Range of Liriodendron chinense
Previous Article in Special Issue
Effects of Linear Disturbances and Fire Severity on Velvet Leaf Blueberry Abundance, Vigor, and Berry Production in Recently Burned Jack Pine Forests
 
 
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

Fire Effects on Historical Wildfire Refugia in Contemporary Wildfires

by
Crystal A. Kolden
1,*,†,
Tyler M. Bleeker
2,†,
Alistair M. S. Smith
1,
Helen M. Poulos
3 and
Ann E. Camp
4
1
Department of Forest, Rangeland, and Fire Sciences, University of Idaho, 875 Perimeter Drive MS 1133, Moscow, ID 83844, USA
2
Department of Geography, University of Idaho, 875 Perimeter Drive MS 3021, Moscow, ID 83844, USA
3
College of the Environment, Wesleyan University, 284 High Street, Middletown, CT 06459, USA
4
School of Forestry and Environmental Studies, Yale University, 195 Prospect Street, New Haven, CT 06511, USA
*
Author to whom correspondence should be addressed.
Equal contribution.
Forests 2017, 8(10), 400; https://doi.org/10.3390/f8100400
Submission received: 2 August 2017 / Revised: 6 October 2017 / Accepted: 10 October 2017 / Published: 20 October 2017
(This article belongs to the Special Issue Wildland Fire, Forest Dynamics, and Their Interactions)

Abstract

:
Wildfire refugia are forest patches that are minimally-impacted by fire and provide critical habitats for fire-sensitive species and seed sources for post-fire forest regeneration. Wildfire refugia are relatively understudied, particularly concerning the impacts of subsequent fires on existing refugia. We opportunistically re-visited 122 sites classified in 1994 for a prior fire refugia study, which were burned by two wildfires in 2012 in the Cascade mountains of central Washington, USA. We evaluated the fire effects for historically persistent fire refugia and compared them to the surrounding non-refugial forest matrix. Of 122 total refugial (43 plots) and non-refugial (79 plots) sites sampled following the 2012 wildfires, one refugial and five non-refugial plots did not burn in 2012. Refugial sites burned more severely and experienced higher tree mortality than non-refugial plots, potentially due to the greater amount of time since the last fire, producing higher fuel accumulation. Although most sites maintained the pre-fire development stage, 19 percent of sites transitioned to Early development and 31 percent of sites converted from Closed to Open canopy. These structural transitions may contribute to forest restoration in fire-adapted forests where fire has been excluded for over a century, but this requires further analysis.

1. Introduction

Wildfire is an integral natural ecosystem process worldwide [1], including the conifer forests of the inland of northwestern United States [2]. In these fire-adapted ecosystems, not every stand is affected equally by wildfire [3,4,5] and differential degrees of ecological change can result from variation in burn severity within single fires [6]. Spatially heterogeneous burn severity drives forest structure and composition [7,8] and produces complexity that is critical to supporting biodiversity, ecosystem services, and forest resilience [2,9,10]. Much of the inland Northwest is characterized by a mixed-severity fire regime that includes both frequent, low-severity fire and infrequent, high-severity fire [2]. Less studied, but equally important ecologically, are forest patches that remain unburned or experience a relatively low degree of change from a wildfire event; these patches are often described as ‘wildfire refugia’ [11,12,13,14]. Such patches represent key landscape elements that support the persistence of fire-sensitive flora and fauna both during the fire (as a refuge) and after the fire (as intact habitat), and provide seed sources for the regeneration of adjacent severely burned areas post-fire [14]. One of the key uncertainties about refugia is their persistence, and what characteristics determine persistent versus temporary refugia during successive ecological disturbances [11]. Resolving this uncertainty is necessary for several reasons, including improving our understanding and the predictability of forest succession dynamics, developing conservation management strategies for critical refugia that are vulnerable, and establishing a baseline from which global change impacts can be measured [14].
Refugia have been broadly defined and delineated, depending on the discipline and research query [15]; to-date, there is no single, widely-accepted definition of fire refugia [4,12,14]. Areas within fire perimeters that do not experience any fire effects, e.g., [16,17], or that experience fire effects at a lower severity than surrounding areas, e.g., [11], have both been described as wildfire refugia. In the literature, the scale and characteristics of wildfire refugia vary by organism and ecosystem [14], leading to different definitions of what constitutes a wildfire refugium. The occurrence of wildfire refugia depends upon several environmental factors that vary spatially and temporally, including topography, climate, soils, geomorphology, and ecological disturbances such as meteorological events, insects, pathogens, and fire [18,19]. These factors interact together to create spatial heterogeneity in vegetation and forest structure, which is maintained by fire [20]. For example, vegetation characteristics such as stand age or structure can either increase or decrease the likelihood of fire occurrence [21,22] or can minimize or exacerbate the effects when a fire does occur [23]. Additionally, topographic complexity influences burn severity [23,24] and refugia formation [11,12]. Generally, bottom-up factors such as vegetation and topography exert a greater influence on burn severity than top-down controls such as weather or climate [24,25,26,27], but in extreme fire weather events, local weather conditions may override vegetative and landscape effects on burn severity [28,29]. Furthermore, human activities such as stand management have the potential to influence the severity of burn patterns and the formation and persistence of refugial patches [30].
Changes in land cover and land use have greatly altered both fire frequency and intensity in inland northwest forests [31], further confounding our ability to understand the formation of fire refugia. Euro-American settlements and associated timber harvesting and grazing, along with a century of fire exclusion, have impacted the forest stand structure [32,33], as well as the spatial distribution and intensity of wildfires, increasing the risk of stand-replacing fire [5,25]. Inland northwest forests are already experiencing climatic conditions conducive to an increased occurrence and duration of wildfires (e.g., extended heat waves and droughts), and these trends are projected to continue through the 21st century [34,35,36,37]. Recent studies suggest that the severity of wildfires may be increasing for some ecosystems [38,39], although the robustness of such trends is questioned by other studies [27,40,41,42]. If the overall fire severity increases, fire refugia that were historically sheltered from fire or experienced only low severity fires might now burn at a higher severity under the contemporary fire regime. To-date, however, there has been little opportunity to assess the effects of the contemporary, altered fire regime on historically persistent fire refugia.
In 2012, the Wenatchee Complex Fires burned through an area of the central Washington state, USA, that had been previously sampled and classified into refugial and non-refugial patches two decades prior to the two fires [11]. This provided a unique opportunity to assess contemporary fire effects on field-delineated refugial and non-refugial stands to examine the persistence of historic wildfire refugia under changing fire regimes and climatic conditions. To our knowledge, no field-based studies have specifically assessed the persistence of wildfire refugia based on the pre-fire identification of long-term wildfire refugia. Camp and colleagues [11] classified historic wildfire refugia using forest stand structure, tree age, and species composition data within the Wenatchee National Forest, defining historic wildfire refugia as forest patches that had been minimally affected by fire events for at least 140 years (well before the onset of fire exclusion or active fire suppression [43]), while the surrounding forest matrix had experienced greater fire effects in historic fire events. The primary research objective of this study was to determine the effects of the 2012 wildfires on these historical fire refugia, specifically their persistence and changes in forest structure. This question was addressed by (1) quantifying and comparing the fire effects between refugial and non-refugial plots as classified pre-fire by Camp et al. [11] and (2) by investigating changes from inferred pre- to post-fire forest stand structure across stands that had been previously classified as refugial or non-refugial. More broadly, this case study evaluates how the occurrence of wildfire refugia and the mosaic of forest structure may shift in response to changing fire regime characteristics in the contemporary era. While conclusions from a case study are limited in application, such studies are needed to identify threats to critical refugia that are central to conservation plans for key species.

2. Materials and Methods

2.1. Study Area

The study area is located in the 47,794 ha Swauk Late Successional Reserve of the Okanogan-Wenatchee National Forest (Cle Elum Ranger District) of central Washington State, USA (Figure 1). A Late Successional Reserve (LSR) is a management designation for an area created through the Northwest Forest Plan with the objective of protecting and enhancing the condition of late-successional and old-growth forest ecosystems; as such, only limited stand management is permitted in LSR-designated areas [44]. Prior to attaining the LSR status in 1995, this area was subject to more intensive management, including selective timber harvesting, clear-cutting, road building, and mining, particularly in lower drainages [11]. The study area is located at the far eastern edge of the Cascade mountain range, extending into the dry interior Columbia River Plateau to the east. Sampled plots in this study ranged in elevation from 1027 to 1912 m. Vegetative communities in the Swauk LSR form a heterogeneous landscape due to strong responses to the dissected topography, precipitation gradient, and insolation differences [45]. At more xeric sites, lower elevations, and south-facing aspects, open-canopy ponderosa pine (Pinus ponderosa Dougl. Ex Forbes) and Douglas-fir (Psuedotsuga menziesii var. menziesii (Mirb.) Franco) stands are common. At more mesic sites, higher elevations, and north-facing slopes, stands of subalpine fir (Abies lasiocarpa (Hook.) Nutt.) and lodgepole pine (Pinus contorta var. latifolia Dougl. ex Loud.) are more typical. However, specific site conditions can cause the immediate juxtaposition of disjunct forest stands. Within the surveyed plots, grand fir (Abies grandis (Dougl. Ex D. Don.)) accounted for 39% of sampled trees, Douglas-fir accounted for 31%, and subalpine fir a further 11%.
The pre-European settlement fire regime near the Swauk LSR varied spatially by vegetative type with a mean fire return interval of seven to 43 years, and with large fires occurring approximately every 27 years [43]. Pre-European settlement fire severity also varied in the study area, with drier forest types experiencing low-severity fire, while more mesic forest types experienced occasional moderate and high-severity fires. However, fire regimes have been significantly altered since European settlement. Fire frequency declined dramatically around 1900, coinciding with the start of commercial logging [43] and the advent of active fire suppression in the Wenatchee National Forest [46]. Consequently, there is no evidence of fire occurrence in the Swauk LSR from 1900 to 2012 [47].

2.2. The 2012 Fires

The Table Mountain and Peavine Canyon (as part of the Wenatchee Complex) fires burned simultaneously after the fires were ignited from lightning strikes in early September 2012. These fires eventually merged, creating a total burn area of 25,274 ha. These fires burned 226 sample sites of the Camp et al. [11] study in three different drainages, where 43 and 183 plots, respectively, were previously classified as refugial and non-refugial. The fires burned under anomalously dry and warm weather conditions compared to 1985 to 2014 climate data recorded at the Swauk Remote Automatic Weather Station (RAWS) approximately 2 km west of the fire perimeter [48]. The average air temperature for the July through September fire season was higher than normal (83rd percentile), and the average air temperature for September when the fires ignited was much higher than normal (93rd percentile). The average relative humidity for the fire season was lower than normal (35th percentile), and the average relative humidity for September was much lower than normal (3rd percentile). Precipitation was slightly below normal for the fire season (38th percentile), but late-season drought was particularly pertinent as the study area went 52 consecutive days without measurable precipitation prior to fire ignition and 34 days without afterwards [48].

2.3. Field Measurements

Field data were collected in the summer and fall of 2014 by resampling the general locations of the Camp et al. [11] sample sites that burned in 2012 (Figure 1). The original Camp et al. [11] plots were established along multiple transects using a hip chain and sighting compass in the field, and the final plot locations were annotated on topographic maps. To determine the global positioning system (GPS) coordinates of these plots, the annotated topographic maps were first digitized into a geographic information system database and then plot coordinates were navigated to with a handheld GPS unit in the field. Since the prior plots were not permanently monumented, new plots were established as close as possible to the original plots within the same forest stand. GPS locations that fell in barren areas or in unsafe field sites were relocated to the nearest suitable forested site. A subset of plots from the Camp et al. [11] study was sampled across the entire elevational gradient for each drainage. A total of 41 refugial sites and 81 non-refugial sites were sampled across the three study drainages for a total of 122 sampled plots.
At each sample site, a modified replication of the Camp et al. [11] sampling protocol was conducted (Figure 2). Plot centers were monumented at the ascribed GPS coordinates and a 15.2-m diameter circular plot covering 725.8 m2 was established. Variables collected or derived for the plot level at each site included four topographic, nine vegetative, and four fire effect variables (Table 1). Plot aspect was measured in degrees and then transformed into northness and eastness indices for analysis. We also collected variables at the tree level for each plot, where a sweeping transect from an azimuth of 0° (north) was used to sample the first ten trees in the plot meeting a minimum threshold diameter at breast height (DBH) of 12.6 cm. We tagged and assessed each set of 10 trees for three demographic and five fire effects variables (Table 2). Three additional vegetative variables at the plot level (average DBH, maximum DBH, and pre-fire plot basal area) were derived from the DBH measurements of the 10 sampled trees on each plot. A total of 1220 trees were sampled in this study.
Burn severity was also assessed at each site using the Composite Burn Index (CBI) protocol [49]. Due to the aggregation of fire effects in the CBI protocol [50], we also assessed the burn severity for each of the five different specific fire effect metrics at the tree level. Per [49], the CBI analysis area was modified to be a 30 by 30-m square to correspond to the size of a Landsat pixel; additionally, the plot was oriented in the cardinal directions to align with the Camp et al. [11] plot azimuths. While CBI is normally conducted one-year post-fire [49], the methodology has been previously utilized to assess fire effects two years post-fire [51]; doing so allows an assessment of longer-term fire effects while also capturing delayed mortality in the tree strata.

2.4. Data Quality Assurance

To ensure that the plots sampled in 2014 were comparable to the plots that Camp et al. [11] originally sampled and classified into potential refugia, we conducted a paired-plot assessment using the original plot data and stand delineations produced by Camp et al. [11]. Stand delineations developed by Camp et al. [11] and aerial imagery were used to determine if the original and resampled plots fell in the same stand. If the original and resampled plot pair was not visually within the same stand, then the topographic and vegetative attributes of each plot were compared for similarities. We were not able to confidently match 13 sampled plots through this qualitative comparison and these were excluded from further paired-plot analysis which resulted in a total of 109 confident paired-plot matches. To confirm that the plots were equivalent to Camp et al. [11], we then used a paired t-test with Welch modification for non-normality to test for differences in the topographic setting of both sample sets [52] by calculating the modified Heat Load Index from McCune and Keon [53], using Equation (1) due to the steep slopes at our study site. As we found no significant differences, we contend that these 109 matched plots provide a conservative match to the Camp et al. [11] plot-level data.

2.5. Comparison of Fire Effects between Resampled and Original Plots

We used the 109 plots determined to be confident matches from the data quality assurance step (Section 2.4) and tested for differences in burn occurrence and burn severity between refugial (n = 36) and non-refugial (n = 73) plots as classified by Camp et al. [11]. Differences in burn occurrence between refugial and non-refugial plots were assessed with a chi-square test of independence.
Fire effects were compared between those plots which Camp et al. [11] classified as refugial versus non-refugial for eight fire effects metrics. Overall plot burn severity was assessed using the Composite Burn Index score and seven individual burn severity metrics (maximum bole char height, percent bole charred, percent foliage scorched, percent foliage torched and percent tree mortality of overstory, and sub-canopy tree strata) were compared between the Camp et al. [11] refugial and non-refugial plots using a Wilcoxon-signed-rank test due to the non-normal distribution of the data (α = 0.1). The Wilcoxon-signed-rank test results in a W-value where lower W-values correspond with lower p-values.

2.6. Assessment of Changes in Forest Structure

To investigate fire impacts on forest structure, we assigned successional states to each plot based on the inferred pre-fire and observed post-fire structural characteristics. In order to assign successional forest structure states to our sample plots, we first needed to assign the type of forested ecosystem in which each plot occurred. A nationwide vegetative community classification called the Biophysical Setting (BpS; NatureServe) was developed as part of the LANDFIRE resource management and planning tool [54,55]. BpS and other LANDFIRE vegetation products are widely used in state-and-transition studies because of the ecologically-based successional states and ecological transitions described in each BpS model [56,57,58,59]. All 122 plots sampled in 2014 were assigned to one of the three most common Biophysical Settings for the study area through an analysis of the 2014 quantitative plot data, recorded qualitative field observations, and photos for each plot. Since each BpS model covers a broad geographical area and cannot account for more localized variation within a single model, BpS models were refined with locally available information on the habitat types of the Wenatchee National Forest [60]; forest types were cross-walked to the Wenatchee National Forest correlate names of the Douglas-fir (n = 17), grand fir (n = 92), and subalpine fir series (n = 13) (Table 3), to be consistent with and comparable to the three forest series reported by Camp et al. [11] in their original analysis.
Once a BpS model and forest series were assigned to each plot, pre- and post-fire forest structure was classified using the Vegetation Dynamics Development Tool (VDDT) [61] classes for each respective BpS model. The VDDT model for each of the three BpS models uses five distinct successional/structural classes (hereafter, referred to as “successional states”): Early Development, Mid-Development Open Canopy, Mid-Development Closed Canopy, Late Development Open Canopy, and Late Development Closed Canopy (Figure 3). Within each BpS model there are different attribute criteria for what constitutes a particular successional state based on species composition; quantitative criteria used in classification included canopy cover, canopy height, maximum tree size class, and average tree size class, while qualitative criteria included the species present, tree relative canopy position, and fuel model (Table 4). Post-fire successional state was classified through post-fire vegetation observed during the 2014 field season. For the pre-fire successional state, in-field estimates of canopy cover and counts of trees presumed living pre-fire were used as best approximations to infer pre-fire vegetative structure and composition. Field notes and plot photos were used to refine this successional state classification when quantitative data alone proved inconclusive.
Once pre-fire and post-fire successional states were assigned to each plot, the transition from one successional state to another due to fire effects was assessed. There were three distinct transitions a plot could have taken due to fire effects (hereafter referred to as “ecological transitions”): (1) plot was maintained in the current successional state; (2) plot canopy was thinned from a closed to open canopy structure of the same development stage; or (3) plot transitioned to an early development successional state. We conducted the transition analysis for all 122 plots (not just the 109 paired plots) as we were not statistically comparing the refugial and non-refugial plots for this analysis, but rather characterizing changes in forest structure.

2.7. Limitations of Methods

As we did not have pre-fire vegetative data, and the Camp et al. [11] plots were not monumented, we inferred pre-fire vegetation attributes with only the burned post-fire vegetation available to sample, which is standard for the CBI protocol [49]. Vegetative reconstruction estimates can be problematic due to the uncertainties of differentiating the magnitude of observed effects solely due to the fire event from the pre-fire conditions and other ecological changes (such as an insect attack or hydrological flow) that occur between the fire and the post-fire measurements [62,63,64]. Canopy cover is particularly difficult to measure from ocular estimates [65], especially after the canopy has been partially consumed in a fire, and it is reasonable to assume that the two-year lag between the fire and our field data collection introduced further error. However, we felt that the opportunity to revisit forested stands that were classified for a refugial objective in a prior study in order to describe fire effects was an opportunity that could not be ignored, despite these limitations.

3. Results

3.1. Assessment of Fire Effects in the Camp et al., Refugial and Non-Refugial Plots

Of the 109 paired plots that were established initially by Camp et al. [11] and that we revisited after the 2012 wildfires, only six (5.5%) did not experience any fire effects in 2012 (i.e., they were unburned). Only one of these six unburned plots was classified as a fire refugium by Camp et al. [11] (Table 5), although this difference in burn occurrence between refugial and non-refugial plots was not statistically significant (chi-square test of independence, χ2 = 0.768, p = 0.381) due to the low number of samples. This translates to <1 percent of plots persisting as refugia and 4.5% having no fire effects and becoming refugia under the most conservative definition.
A comparison of the fire effects between the 109 matched Camp et al. [11] refugial and non-refugial plots revealed a trend in differences in burn severity, where classified refugial plots generally experienced greater fire effects (Figure 4). Refugial plots burned more severely than non-refugial plots as assessed through the total plot CBI metric (W = 1015.5, p = 0.0582). For the tree-level severity metrics, the percent total tree mortality (W = 921.5, p = 0.0196), percent understory tree mortality (W = 867, p = 0.0116), average bole char height (W = 1042, p = 0.0802), and average foliage scorched (W = 899, p = 0.00754) were all significantly higher for refugial plots (at α = 0.1, used due to non-normal data and a greater number of non-refugial plots). Percent overstory tree mortality and percent bole char observations were also higher for refugial plots, but these differences were not statistically significant. Foliage torched showed no significant difference between refugial and non-refugial plots, but this result is likely attributable to low overall levels of foliar torching, with the exception of a few highly torched plots resulting from crown fires.

3.2. Changes in Successional State

Based on our reconstruction of forest structure and successional state pre-fire, there were no Early Development successional state plots pre-fire, with most of the plots classified at Mid-Development (32% Open Canopy and 39% Closed Canopy), and the remaining 29% classified as Late Development (9% Open Canopy and 20% Closed Canopy) (Table 6). Post-fire, 19% of sample plots were classified as an Early Development successional state, 26% were classified as Late Development, and most plots were still Mid-Development (55%). Half (50%) of the sampled plots did not change in terms of the successional state due to fire effects, while nearly one-third (31%) transitioned from a closed to open-canopy structure (for both Mid- and Late Development categories combined). Closed Canopy plots classified as either Mid- or Late Development successional state (59%) were more abundant than Open Canopy plots pre-fire (41%), but Open Canopy plots (65%) were four times more abundant than Closed Canopy plots (16%) post-fire.
Assessing structural transitions by whether plots were classified as refugial or non-refugial by Camp et al. [11] reveals that a higher proportion of pre-fire refugia transitioned to Early development successional state (24%) than non-refugial plots (16%) (Table 7). More refugia were also thinned by the fire from Closed to Open Canopy (34%; in both Mid and Late-development states combined) than non-refugia (30%). Accordingly, more non-refugial plots maintained their pre-fire successional state.

4. Discussion

4.1. Comparison of Fire Effects between Sample Years

We found that the plots classified by Camp et al. [11] as refugial experienced more severe fire effects from the 2012 wildfires in comparison to classified non-refugial plots. This finding supports an inference of Camp et al. [11] in their study; they noted that pre-European settlement fire refugia appeared to have higher fire intensities and severities associated with longer fire return intervals. What was particularly surprising in the present study was just how few of the 122 plots sampled were unburned (<6%). Studies quantifying the unburned proportion across entire fires at local to regional scales have found a very broad range of the proportion unburned for individual fires [4,42], with Meddens et al. [13] reporting a regional average of 20% unburned for the inland Northwest. As these plots were specifically located by Camp et al. [11] to capture prospective fire refugia, the low proportion of unburned compared to regional averages further supports their conclusion that refugial plots, when they do eventually burn, do so with a higher severity than the non-refugial surrounding matrix. Because of the longer fire return intervals in refugial patches, tree species with thin bark and minimal self-pruning, such as grand fir, are able to establish and develop as ladder fuels that facilitate the surface fire ignition of crown fuels due to their lower canopy base height [32], a concept that Camp et al. [11] termed ‘outgrowing’ the refugial status. The additional fuels accumulated in these ‘outgrown’ refugial patches then lead such sites to burn at a higher severity than the surrounding matrix.

4.2. Distribution of Post-Fire Successional States and Ecological Transitions

We found that less than 30 percent of plots resampled in 2014 were Late Development stage pre-fire, with Mid-Development stage plots being the most common. This successional composition is relatively consistent with the pre-European settlement fire regime [39], although occasional high severity patches of fire (as part of the mixed-fire regime) would have also produced a few Early Development patches scattered across the landscape during that period. Here, however, we encountered no plots that were Early Development successional state pre-fire, which was unsurprising given that prior to 2012, the most recent wildfire in the study area occurred before 1900 due to the effectiveness of US fire suppression policies in the 20th century [46]. Post-fire, 19 percent of sampled plots transitioned to the Early Development state, increasing the structural heterogeneity of the study area [66]. Open Canopy stands were four times more abundant than Closed Canopy stands post-fire, whereas Closed Canopy stands were more abundant pre-fire. These sorts of transitions are consistent with the restoration needs highlighted by Haugo et al. [59] for these forests, and suggest that in this case study, at least, the fire behavior may have played a restorative role by creating both early successional openings and a more open canopy from a previously closed canopy.
The relationship between the more severe fire effects and the forest structural changes becomes evident when stratifying structural changes by whether plots were classified as refugia pre-fire. This fire-induced transition to an increased open canopy structure is consistent with mixed severity fires reducing canopy closure [32,43,67], and occurred in a higher proportion of refugial plots because they burned at a higher severity. Similarly, more refugial plots transitioned to Early Development successional state due to a higher burn severity in refugial plots. One point of interest in this breakdown is the number of plots that maintained their successional state (as there were no Early state plots pre-fire, these were all Mid- and Late-Development sites). Most of these plots (all but six) experienced some fire effects; these effects were of a low enough severity to maintain the forest structure successional state given the definitions in the BpS models.

4.3. Implications for Management

In the face of environmental change, managers can take many actions to increase ecosystem resilience [36,68]. Many plant and animal species are fire-sensitive and require refugial habitats to persist in the landscape [2], including species of high management concern such as the northern spotted owl (Strix occidentalis caurina), which was the focal species of the Camp et al. [11] refugia analysis. There is much concern over the persistence of refugia given climate change and changing ecological disturbance regimes [41], but our results demonstrate that even with fire re-entry into sites with over a century of fire exclusion, some refugial plots still maintained pre-fire forest structure and some previously non-refugial plots served as during-fire refugia in the 2012 fires. For managers seeking to conserve refugia for given species, there are likely pathways to doing so that do not require the total exclusion of wildfire, but additional research is needed to improve quantitative landscape composition models and determine when forests have become vulnerable to permanent structural transitions that can eradicate refugia [69].

5. Conclusions

Wildfire refugia in forested landscapes are critical to species survivorship both during and between fires, as well as facilitating forest regeneration in adjacent burned areas. However, refugia research is still nascent, with the definitions and delineation of refugia being highly variable and dependent upon species of interest. This case study provided a unique opportunity to assess the effects of a wildfire on fire refugia in central Washington State, USA, that were classified and delineated by Camp et al. [11] over two decades ago and subsequently burned in two 2012 wildfires. Our findings that almost all of the plots burned, but half of the plots persisted in their pre-fire forest structure successional state, suggests that definitions of fire refugia focused on the maintenance of forest structure or canopy thresholds may reveal a higher proportion of refugia that are persistent through multiple fires. We also found that nearly one-fifth of the plots converted to Early Development state due to a high burn severity, and almost one-third experienced enough crown loss to transition from Closed to Open Canopy; these types of transitions are critical to forest restoration following over a century of fire exclusion in forested landscapes across the western US. Ultimately, this opportunistic case study highlights that the re-entry of fire into forests where fire has been excluded produces a range of fire effects that include the maintenance of pre-fire forest structure. Additionally, many current fire refugia may be vulnerable to future fire, particularly where fire exclusion has allowed for fuel accumulation and the growth of fire-sensitive species. Refugia more broadly, however, require much additional study to improve our understanding of their persistence, vulnerability, and role in forest structural dynamics.

Acknowledgments

Arjan Meddens, Phil Higuera, and Paul Hessburg contributed insight to the project and helped with data analysis. Additional thanks to Ryan McCarley for help in field data collection and figure editing. Support for this research came from NASA Award NNX10AT77A, NSF Idaho EPSCoR under award EPS-0814387, the USDA Forest Service Western Wildlands Environmental Threat Assessment Center under agreement number 13-JV-11261900-072, and the USGS Northwest Climate Science Center award G14AP00177. Smith was partially supported under NASA Award NNX11AO24G. This work was partially supported by the National Science Foundation under award no. DMS-1520873.

Author Contributions

C.A.K., T.M.B. and H.M.P. conceived and designed the experiments; T.M.B. performed the experiments; H.M.P. and A.E.C. contributed historical data and aerial photos; C.A.K. and T.M.B. analyzed the data; all authors wrote the paper.

Conflicts of Interest

The authors declare no conflict of interest.

References

  1. Bowman, D.M.; Balch, J.K.; Artaxo, P.; Bond, W.J.; Carlson, J.M.; Cochrane, M.A.; D’Antonio, C.M.; DeFries, R.S.; Doyle, J.C.; Harrison, S.P.; et al. Fire in the Earth system. Science 2009, 324, 481–484. [Google Scholar] [CrossRef] [PubMed]
  2. Agee, J.K. Fire Ecology of Pacific Northwest Forests; Island Press: Washington, DC, USA, 1996. [Google Scholar]
  3. Lentile, L.B.; Morgan, P.; Hudak, A.T.; Bobbitt, M.J.; Lewis, S.A.; Smith, A.M.S.; Robichaud, P. Post-fire burn severity and vegetation response following eight large wildfires across the western United States. Fire Ecol. 2007, 3, 91–108. [Google Scholar] [CrossRef]
  4. 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]
  5. Reilly, M.J.; Dunn, C.J.; Meigs, G.W.; Spies, T.A.; Kennedy, R.E.; Bailey, J.D.; Briggs, K. Contemporary patterns of fire extent and severity in forests of the Pacific Northwest, USA (1985–2010). Ecosphere 2017, 8. [Google Scholar] [CrossRef]
  6. Lentile, L.B.; Holden, Z.A.; Smith, A.M.S.; Falkowski, M.J.; Hudak, A.T.; Morgan, P.; Lewis, S.A.; Gessler, P.E.; Benson, N.C. Remote sensing techniques to assess active fire characteristics and post-fire effects. Int. J. Wildland Fire 2006, 15, 319–345. [Google Scholar] [CrossRef]
  7. Hessburg, P.F.; Salter, R.B.; James, K.M. Re-examining fire severity relations in pre-management era mixed conifer forests: Inferences from landscape patterns of forest structure. Landsc. Ecol. 2007, 22, 5–24. [Google Scholar] [CrossRef]
  8. Freund, J.A.; Franklin, J.F.; Lutz, J.A. Structure of early old-growth Douglas-fir forests in the Pacific Northwest. For. Ecol. Manag. 2015, 335, 11–25. [Google Scholar] [CrossRef]
  9. Hutto, R.L. The ecological importance of severe wildfires: Some like it hot. Ecol. Appl. 2008, 18, 1827–1834. [Google Scholar] [CrossRef] [PubMed]
  10. Vaillant, N.M.; Kolden, C.A.; Smith, A.M.S. Assessing landscape vulnerability to wildfire in the United States. Curr. For. Rep. 2016, 2, 201–213. [Google Scholar] [CrossRef]
  11. Camp, A.; Oliver, C.; Hessburg, P.; Everett, R. Predicting late-successional fire refugia pre-dating European settlement in the Wenatchee Mountains. For. Ecol. Manag. 1997, 95, 63–77. [Google Scholar] [CrossRef]
  12. Krawchuk, M.A.; Haire, S.L.; Coop, J.; Parisien, M.A.; Whitman, E.; Chong, G.; Miller, C. Topographic and fire weather controls of fire refugia in forested ecosystems of northwestern North America. Ecosphere 2016, 7. [Google Scholar] [CrossRef]
  13. Meddens, A.J.H.; Kolden, C.A.; Lutz, J.A. Detecting unburned areas within wildfire perimeters using Landsat and ancillary data across the northwestern United States. Remote Sens. Environ. 2016, 186, 275–285. [Google Scholar] [CrossRef]
  14. Robinson, N.M.; Leonard, S.W.; Ritchie, E.G.; Bassett, M.; Chia, E.K.; Buckingham, S.; Gibb, H.; Bennett, A.F.; Clarke, M.F. Refuges for fauna in fire-prone landscapes: Their ecological function and importance. J. Appl. Ecol. 2013, 50, 1321–1329. [Google Scholar] [CrossRef]
  15. Keppel, G.; Van Niel, K.P.; Wardell-Johnson, G.W.; Yates, C.J.; Byrne, M.; Mucina, L.; Schut, A.G.; Hopper, S.D.; Franklin, S.E. Refugia: Identifying and understanding safe havens for biodiversity under climate change. Glob. Ecol. Biog. 2012, 21, 393–404. [Google Scholar] [CrossRef]
  16. Delong, S.C.; Kessler, D.W. Ecological characteristics of mature forest remnants left by wildfire. For. Ecol. Manag. 2000, 131, 93–106. [Google Scholar] [CrossRef]
  17. Schwilk, D.W.; Keeley, J.E. The role of wildfire refugia in the distribution of Pinus sabiniana (Pinaceae) in the southern Sierra Nevada. Madroño 2006, 53, 364–372. [Google Scholar] [CrossRef]
  18. Román-Cuesta, R.M.; Garcia, M.; Retana, J. Factors influencing the formation of unburned forest islands within the perimeter of a large forest fire. For. Ecol. Manag. 2009, 258, 71–80. [Google Scholar] [CrossRef]
  19. Stine, P.; Hessburg, P.; Spies, T.; Kramer, M.; Fettig, C.J.; Hansen, A.; Lehmkuhl, J.; O’Hara, K.; Polivka, K.; Singleton, P.; et al. The Ecology and Management of Moist Mixed-Conifer Forests in Eastern Oregon and Washington: A Synthesis of the Relevant Biophysical Science and Implications for Future Land Management; General Technical Report; PNW-GTR-897; United States Department of Agriculture Forest Service: Portland, OR, USA, 2014.
  20. Kane, V.R.; Lutz, J.A.; Roberts, S.L.; Smith, D.F.; McGaughey, R.J.; Povak, N.A.; Brooks, M.L. Landscape-scale effects of fire severity on mixed-conifer and red fir forest structure in Yosemite National Park. For. Ecol. Manag. 2013, 287, 17–31. [Google Scholar] [CrossRef]
  21. Kushla, J.D.; Ripple, W.J. The role of terrain in a fire mosaic of a temperate coniferous forest. For. Ecol. Manag. 1997, 95, 97–107. [Google Scholar] [CrossRef]
  22. Alexander, J.D.; Seavy, N.E.; Ralph, C.J.; Hogoboom, B. Vegetation and topographical correlates of fire severity from two fires in the Klamath-Siskiyou region of Oregon and California. Int. J. Wildland Fire 2006, 15, 237. [Google Scholar] [CrossRef]
  23. Kane, V.R.; Cansler, C.A.; Povak, N.A.; Kane, J.T.; McGaughey, R.J.; Lutz, J.A.; Churchill, D.J.; North, M.P. Mixed severity fire effects within the Rim fire: Relative importance of local climate, fire weather, topography, and forest structure. For. Ecol. Manag. 2015, 358, 62–79. [Google Scholar] [CrossRef]
  24. Dillon, G.K.; Holden, Z.A.; Morgan, P.; Crimins, M.A.; Heyerdahl, E.K.; Luce, C.H. Both topography and climate affected forest and woodland burn severity in two regions of the western US, 1984–2006. Ecosphere 2011, 2, 130. [Google Scholar] [CrossRef]
  25. Cansler, C.A.; McKenzie, D. Climate, fire size, and biophysical setting control fire severity and spatial pattern in the northern Cascade Range, USA. Ecol. Appl. 2014, 24, 1037–1056. [Google Scholar] [CrossRef] [PubMed]
  26. Birch, D.S.; Morgan, P.; Kolden, C.A.; Abatzoglou, J.T.; Dillon, G.K.; Hudak, A.T.; Holden, Z.A.; Smith, A.M.S. Vegetation, topography, and daily weather influenced burn severity in central Idaho and western Montana forests. Ecosphere 2015, 6, 17. [Google Scholar] [CrossRef]
  27. Abatzoglou, J.T.; Kolden, C.A.; Williams, A.P.; Lutz, J.A.; Smith, A.M.S. Climatic influences on inter-annual variability in regional burn severity across western US forests. Int. J. Wildland Fire 2017, 26, 269–275. [Google Scholar] [CrossRef]
  28. Bessie, W.C.; Johnson, E.A. Relative importance of fuels and weather on fire behavior in subalpine forests. Ecology 1995, 76, 747–762. [Google Scholar] [CrossRef]
  29. Birch, D.S.; Morgan, P.; Kolden, C.A.; Hudak, A.T.; Smith, A.M.S. Is proportion burned severely related to daily area burned? Environ. Res. Lett. 2014, 9, 064011. [Google Scholar] [CrossRef]
  30. Prichard, S.J.; Kennedy, M.C. Fuel treatments and landform modify landscape patterns of burn severity in an extreme fire event. Ecol. Appl. 2014, 24, 571–590. [Google Scholar] [CrossRef] [PubMed]
  31. Hessburg, P.F.; Agee, J.K. An environmental narrative of Inland Northwest United States forests, 1800–2000. For. Ecol. Manag. 2003, 178, 23–59. [Google Scholar] [CrossRef]
  32. Hessburg, P.F.; Agee, J.K.; Franklin, J.F. Dry forests and wildland fires of the inland Northwest USA: Contrasting the landscape ecology of the pre-settlement and modern eras. For. Ecol. Manag. 2005, 211, 117–139. [Google Scholar] [CrossRef]
  33. Hessburg, P.F.; Smith, B.G.; Salter, R.B.; Ottmar, R.D.; Alvarado, E. Recent changes (1930s–1990s) in spatial patterns of interior northwest forests, USA. For. Ecol. Manag. 2000, 136, 53–83. [Google Scholar] [CrossRef]
  34. Littell, J.S.; McKenzie, D.; Peterson, D.L.; Westerling, A.L. Climate and wildfire area burned in western U.S. ecoprovinces, 1916–2003. Ecol. Appl. 2009, 19, 1003–1021. [Google Scholar] [CrossRef] [PubMed]
  35. Abatzoglou, J.T.; Kolden, C.A. Relationships between climate and macroscale area burned in the western United States. Int. J. Wildland Fire 2013, 22, 1003–1020. [Google Scholar] [CrossRef]
  36. Barbero, R.; Abatzoglou, J.T.; Larkin, N.K.; Kolden, C.A.; Stocks, B. Climate change presents increased potential for very large fires in the contiguous United States. Int. J. Wildland Fire 2015, 24, 892–899. [Google Scholar] [CrossRef]
  37. Abatzoglou, J.T.; Williams, A.P. Impact of anthropogenic climate change on wildfire across western US forests. Proc. Nat. Acad. Sci. USA 2016, 113, 11770–11775. [Google Scholar] [CrossRef] [PubMed]
  38. Miller, J.D.; Safford, H.D.; Crimmins, M.; Thode, A.E. Quantitative evidence for increasing forest fire severity in the Sierra Nevada and southern Cascade mountains, California and Nevada, USA. Ecosystems 2008, 12, 16–32. [Google Scholar] [CrossRef]
  39. Miller, J.D.; Safford, H. Trends in wildfire severity: 1984 to 2010 in the Sierra Nevada, Modoc Plateau, and southern Cascades, California, USA. Fire Ecol. 2012, 8, 41–57. [Google Scholar] [CrossRef]
  40. Picotte, J.J.; Peterson, B.; Meier, G.; Howard, S.M. 1984–2010 trends in fire burn severity and area for the conterminous US. Int. J. Wildland Fire 2016, 25, 413–420. [Google Scholar] [CrossRef]
  41. 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 2017, in press. [Google Scholar]
  42. 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]
  43. Wright, C.S.; Agee, J.K. Fire and vegetation history in the eastern Cascade Mountains, Washington. Ecol. Appl. 2004, 14, 443–459. [Google Scholar] [CrossRef]
  44. United States Department of Agriculture (USDA); United States Department of the Interior (USDOI). Record of Decision for Amendments to Forest Service and Bureau of Land Management Planning Documents within the Range of the Northern Spotted Owl, Attachment A: Standards and Guidelines for Management of Habitat for Late-Successional and Old-Growth Forest Related Species within the Range of the Northern Spotted Owl; U.S. Government: Washington, DC, USA, 1994.
  45. Williams, C.K.; Smith, B. Forested Plant Associations of the Wenatchee National Forest; US Forest Service Pacific Northwest Region, Wenatchee National Forest: Portland, OR, USA, 1991.
  46. Holstine, C.E. An Historical Overview of the Wenatchee National Forest, Washington; Rep. 100–80; Archaeological and Historical Services, Eastern Washington University: Cheney, WA, USA, 1992. [Google Scholar]
  47. Everett, R.L.; Martin, S.; Bickford, M.; Schellhaas, R.; Forsman, E. Variability and Dynamics of Spotted Owl Nesting Habitat in Eastern Washington; General Technical Report; INT-GTR-291; United States Department of Agriculture Forest Service: Ogden, UT, USA, 1991; pp. 35–39.
  48. Western Regional Climate Center. Available online: www.wrcc.dri.edu (accessed on 21 March 2015).
  49. Key, C.H.; Benson, N.C. Landscape Assessment (LA) Sampling and Analysis Methods; US Forest Service: Ogden, UT, USA.
  50. Morgan, P.; Keane, R.E.; Dillon, G.K.; Jain, T.B.; Hudak, A.T.; Karau, E.C.; Sikkink, P.G.; Holden, Z.A.; Strand, E.K. Challenges of assessing fire and burn severity using field measures, remote sensing and modelling. Int. J. Wildland Fire 2014, 23, 1045–1060. [Google Scholar] [CrossRef]
  51. Zhu, Z.; Key, C.H.; Ohlen, D.; Benson, N.C. Evaluating Sensitivities of Burn Severity-Mapping Algorithms for Different Ecosystems and Fire Histories; Joint Fire Science Program: Boise, ID, USA, 2006.
  52. Welch, B.L. The generalization of “Student’s” problem when several different population variances are involved. Biometrika 1947, 34, 28–35. [Google Scholar] [CrossRef] [PubMed]
  53. McCune, B.; Keon, D. Equations for potential annual direct incident radiation and heat load. J. Veg. Sci. 2002, 13, 603–606. [Google Scholar] [CrossRef]
  54. LANDFIRE. Available online: www.landfire.gov (accessed on 24 September 2015).
  55. Rollins, M.G. LANDFIRE: A nationally consistent vegetation, wildland fire, and fuel assessment. Int. J. Wildland Fire 2009, 18, 235–249. [Google Scholar] [CrossRef]
  56. Keane, R.E.; Burgan, R.; van Wagtendonk, J.W. Mapping wildland fuels for fire management across multiple scales: Integrating remote sensing, GIS, and biophysical modeling. Int. J. Wildland Fire 2001, 10, 301–319. [Google Scholar] [CrossRef]
  57. Keane, R.E.; Karau, E.C. Evaluating the ecological benefits of wildfire by integrating fire and ecosystem simulation models. Ecol. Model. 2010, 221, 1162–1172. [Google Scholar] [CrossRef]
  58. Strand, E.K.; Vierling, L.A.; Bunting, S.C.; Gessler, P.E. Quantifying successional rates in western aspen woodlands: Current conditions, future predictions. For. Ecol. Manag. 2009, 257, 1705–1715. [Google Scholar] [CrossRef]
  59. Haugo, R.; Zanger, C.; DeMeo, T.; Ringo, C.; Shlisky, A.; Blankenship, K.; Simpson, M.; Mellen-McLean, K.; Kertis, J.; Stern, M. A new approach to evaluate forest structure restoration needs across Oregon and Washington. For. Ecol. Manag. 2015, 335, 37–50. [Google Scholar] [CrossRef]
  60. Lillybridge, T.R.; Kovalchik, B.L.; Williams, C.K.; Smith, B.G. Field Guide for Forested Plant Associations of the Wenatchee National Forest; Gen. Tech. Rep. 359; US Forest Service Pacific Northwest Research Station: Portland, OR, USA, 1995. [Google Scholar]
  61. ESSA Technologies, Ltd. Vegetation Dynamics Development Tool User Guide, Version 6.0; ESSA Technologies Ltd.: Vancouver, BC, Canada, 2007. [Google Scholar]
  62. Smith, A.M.S.; Eitel, J.U.H.; Hudak, A.T. Spectral Analysis of Charcoal on Soils: Implications for Wildland Fire Severity Mapping Methods. Int. J. Wildland Fire 2010, 19, 976–983. [Google Scholar] [CrossRef]
  63. Roy, D.P.; Boschetti, L.; Smith, A.M.S. Satellite remote sensing of fires. In Fire Phenomena and the Earth System: An Interdisciplinary Guide to Fire Science; Belcher, C.M., Rein, G., Eds.; John Wiley & Sons Ltd.: Chichester, UK, 2013; ISBN 978-0-470-65748-5. [Google Scholar]
  64. Smith, A.M.S.; Sparks, A.M.; Kolden, C.A.; Abatzoglou, J.T.; Talhelm, A.F.; Johnson, D.M.; Boschetti, L.; Lutz, J.A.; Apostol, K.G.; Yedinak, K.M.; et al. Towards a new paradigm in fire severity research using dose-response experiments. Int. J. Wildland Fire 2016, 25, 158–166. [Google Scholar] [CrossRef]
  65. Korhonen, L.; Korhonen, K.T.; Rautiainen, M.; Stenberg, P. Estimation of forest canopy cover: A comparison of field measurement techniques. Silva Fennica 2006, 40, 577–588. [Google Scholar] [CrossRef]
  66. Swanson, M.E.; Franklin, J.F.; Beschta, R.L.; Crisafulli, C.M.; DellaSala, D.A.; Hutto, R.L.; Lindenmayer, D.B.; Swanson, F.J. The forgotten stage of forest succession: Early-successional ecosystems on forest sites. Front. Ecol. Environ. 2011, 9, 117–125. [Google Scholar] [CrossRef]
  67. Barrett, S. Fire suppression’s effects on forest succession within a central Idaho wilderness. West. J. Appl. For. 1988, 3, 76–80. [Google Scholar]
  68. Hessburg, P.F.; Churchill, D.J.; Larson, A.J.; Haugo, R.D.; Miller, C.; Spies, T.A.; North, M.P.; Povak, N.A.; Belote, R.T.; Singleton, P.H.; et al. Restoring fire-prone Inland Pacific landscapes: Seven core principles. Landsc. Ecol. 2015, 30, 1805–1835. [Google Scholar] [CrossRef]
  69. Smith, A.M.S.; Kolden, C.A.; Tinkham, W.T.; Talhelm, A.F.; Marshall, J.D.; Hudak, A.T.; Boschetti, L.; Falkowski, M.J.; Greenberg, J.A.; Anderson, J.W.; et al. Remote sensing the vulnerability of vegetation in natural terrestrial ecosystems. Remote Sens. Environ. 2014, 154, 322–337. [Google Scholar] [CrossRef]
Figure 1. Study area in the east Cascades of central Washington State, USA, with burn perimeters of the 2012 Peavine Canyon and Table Mountain fires and spatial locations of the Camp et al. [11] classified refugial (diamonds) and non-refugial (circles) plots in the three study drainages (Tronsen, Mission, and Boulder).
Figure 1. Study area in the east Cascades of central Washington State, USA, with burn perimeters of the 2012 Peavine Canyon and Table Mountain fires and spatial locations of the Camp et al. [11] classified refugial (diamonds) and non-refugial (circles) plots in the three study drainages (Tronsen, Mission, and Boulder).
Forests 08 00400 g001
Figure 2. Diagram of the 2014 field sampling protocol.
Figure 2. Diagram of the 2014 field sampling protocol.
Forests 08 00400 g002
Figure 3. State-and-transition model of the Vegetation Dynamics Development Tool successional states and ecological transitions as affected by differential wildfire severity (modified from [4]).
Figure 3. State-and-transition model of the Vegetation Dynamics Development Tool successional states and ecological transitions as affected by differential wildfire severity (modified from [4]).
Forests 08 00400 g003
Figure 4. Comparison of fire effects between Camp et al. [11] refugial and non-refugial plots for eight different burn severity metrics. * p ≤ 0.10, ** p ≤ 0.05, *** p ≤ 0.01.
Figure 4. Comparison of fire effects between Camp et al. [11] refugial and non-refugial plots for eight different burn severity metrics. * p ≤ 0.10, ** p ≤ 0.05, *** p ≤ 0.01.
Forests 08 00400 g004
Table 1. Plot-level topographic, vegetative, and fire effects variables.
Table 1. Plot-level topographic, vegetative, and fire effects variables.
Data TypeVariableDefinition
TopographicAspect Northnesscos(π/2-aspect); range from −1 (south) to 1 (north)
Aspect Eastnesssin(π/2-aspect); range from −1 (west) to 1 (east)
SlopeDegrees; measured by clinometer
ElevationMeters; measured by global positioning system
Topography TypeTen option categorical classification (from [11])
VegetativeMax Canopy HeightMeters; measured by Impulse Laser
Species PresentDominant tree species
Canopy StructurePresence/absence of overstory/subcanopy strata
All Trees Pre-fireCount of all trees alive in plot pre-fire
Overstory Trees Pre-fireCount of all ≥12.6 cm DBH alive in overstory strata pre-fire
Subcanopy Trees Pre-fireCount of all trees ≥12.6 cm DBH alive in subcanopy strata pre-fire
Total Canopy Cover Pre-fireOcular estimate of pre-fire canopy cover for all tree strata
Overstory Canopy Cover Pre-fireOcular estimate of pre-fire canopy cover for overstory strata
Subcanopy Canopy Cover Pre-fireOcular estimate of pre-fire canopy cover for subcanopy strata
Average DBHAverage DBH of 10 sampled trees on plot
Maximum DBHMaximum DBH of 10 sampled trees on plot
Pre-fire Plot Basal AreaAverage basal area of 10 sampled trees on plot [π(DBH/2)2] * count of trees on plot. Unit: m2 basal area/900 m2 plot area
Fire EffectsTotal Tree MortalityCount of all tree mortality in plot post-fire
Overstory Tree MortalityCount of all ≥12.6 cm DBH tree mortality in overstory strata post-fire
Subcanopy Tree MortalityCount of all ≥12.6 cm DBH subcanopy tree mortality in subcanopy strata post-fire
Total Plot CBIComposite Burn Index protocol (score from 0 to 3)
DBH—diameter at breast height; CBI—Composite Burn Index [49].
Table 2. Tree-level demographic and fire effect variables.
Table 2. Tree-level demographic and fire effect variables.
Data TypeVariableDefinition
DemographicSpeciesField Identification
Diameter at breast heightField measure; cm
Secondary StressPresence of: Fire, Freezing, Fungus, Insect, Mechanical, Mistletoe, Rot
Fire EffectsMortalityFire-induced tree death
Percent Bole CharMaximum percent of basal bole with visible char
Bole Char Max HeightMaximum height of continuous char on bole (m)
Percent Foliage ScorchOcular estimate of pre-fire living foliage scorched or girdled
Percent Foliage TorchOcular estimate of pre-fire living foliage torched by fire
Table 3. Cross-walk between the BpS Model and Wenatchee NF Correlate from [60].
Table 3. Cross-walk between the BpS Model and Wenatchee NF Correlate from [60].
Biophysical Setting Model NameWenatchee NF Correlate Plots
Northern Rocky Mountain Dry-Mesic Montane Mixed Conifer ForestDouglas-fir Series17
East Cascades Mesic Montane Mixed-Conifer Forest and WoodlandGrand Fir Series84
Rocky Mountain Subalpine Dry-Mesic Spruce-Fir Forest and WoodlandSubalpine Fir Series13
Table 4. Quantitative criteria and thresholds utilized to classify plots into the five successional states that describe the forest structure for the three Biophysical Setting (BpS) models present in the study area. Both the initial classification of the plots into BpS models and further classification into successional states also utilized qualitative information and species lists in the BpS model descriptions, which can be found on the LANDFIRE website (www.landfire.gov).
Table 4. Quantitative criteria and thresholds utilized to classify plots into the five successional states that describe the forest structure for the three Biophysical Setting (BpS) models present in the study area. Both the initial classification of the plots into BpS models and further classification into successional states also utilized qualitative information and species lists in the BpS model descriptions, which can be found on the LANDFIRE website (www.landfire.gov).
Biophysical Setting (BpS)StateCoverHeightTree Size Class
Northern Rocky Mountain Dry-Mesic Montane Mixed Conifer Forest (Douglas Fir Series)Early0–20%Tree 0–5 mSapling > 4.5’, <5” DBH
Mid-Open0–40%Tree 5.1–25 m9–21” DBH
Mid-Closed41–100%Tree 5.1–25 m9–21” DBH
Late-Open11–40%Tree 25.1–50 m > 33” DBH
Late-Closed41–100%Tree 25.1–50 m > 33” DBH
East Cascades Mesic Montane Mixed Conifer Forest and Woodland (Grand Fir series)Early0–100%Tree 0–10 mSapling >4.5’, <5” DBH
Mid-Open0–60%Tree 10.1–25 m9–21” DBH
Mid-Closed61–100%Tree 10.1–25 m9–21” DBH
Late-Open0–60%Tree 25.1–>50 m >33” DBH
Late-Closed61–100%Tree 25.1–>50 m >33” DBH
Rocky Mountain Subalpine Dry-Mesic Spruce-Fir Forest and Woodland (Subalpine Fir series)Early0–40%Shrub 0–0.5 mNone
Mid-Open11–30%Tree 5.1–10 m9–21” DBH
Mid-Closed31–60%Tree 5.1–10 m9–21” DBH
Late-Open11–40%Tree 10.1–25 m21–33” DBH
Late-Closed41–70%Tree 10.1–25 m21–33” DBH
Table 5. Number of plots classified in the original Camp et al. [11] study that burned in the 2012 fires, re-surveyed in the present study, paired to the plots from [11], and found to have no fire effects (i.e., the most conservative definition of refugial) by this study.
Table 5. Number of plots classified in the original Camp et al. [11] study that burned in the 2012 fires, re-surveyed in the present study, paired to the plots from [11], and found to have no fire effects (i.e., the most conservative definition of refugial) by this study.
Sampling DescriptionRefugial per [11]Non-Refugial per [11]Total
Sampled by [11]43183226
Sampled in 2014, this study4181122
Plots paired to [11]3673109
Plots unburned156
Table 6. Distribution of successional states for all 122 sampled plots according to pre-fire and post-fire successional state for all Biophysical Setting models.
Table 6. Distribution of successional states for all 122 sampled plots according to pre-fire and post-fire successional state for all Biophysical Setting models.
Successional StatePre-Fire StatePost-Fire State
CountPercentCountPercent
Early Development00%2319%
Mid-Development Open Canopy3932%5545%
Mid-Development Closed Canopy4739%1210%
Late Development Open Canopy119%2420%
Late Development Closed Canopy2520%86%
Table 7. Primary plot structural transitions by pre-fire refugia classification.
Table 7. Primary plot structural transitions by pre-fire refugia classification.
Successional TransitionRefugiaNon-Refugia
CountPercentCountPercent
Maintained state1742%4454%
Thinned from Closed to Open Canopy1434%2430%
Converted to Early Development state1024%1316%

Share and Cite

MDPI and ACS Style

Kolden, C.A.; Bleeker, T.M.; Smith, A.M.S.; Poulos, H.M.; Camp, A.E. Fire Effects on Historical Wildfire Refugia in Contemporary Wildfires. Forests 2017, 8, 400. https://doi.org/10.3390/f8100400

AMA Style

Kolden CA, Bleeker TM, Smith AMS, Poulos HM, Camp AE. Fire Effects on Historical Wildfire Refugia in Contemporary Wildfires. Forests. 2017; 8(10):400. https://doi.org/10.3390/f8100400

Chicago/Turabian Style

Kolden, Crystal A., Tyler M. Bleeker, Alistair M. S. Smith, Helen M. Poulos, and Ann E. Camp. 2017. "Fire Effects on Historical Wildfire Refugia in Contemporary Wildfires" Forests 8, no. 10: 400. https://doi.org/10.3390/f8100400

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Metrics

Back to TopTop