Effects of Megafire on Woody Species in the Mixed-Grass Prairie
Matthias W. Sirch1*, Daniel S. Sullins2, Nicholas J. Parker2, David A. Haukos3, John D. Kraft1, Christian A. Hagen4, and Kent A. Fricke5
1Kansas Cooperative Fish and Wildlife Research Unit, Division of Biology, Kansas State University, 205 Leasure Hall, 1128 N. 17th Street, Manhattan, KS 66506, USA. 2Department of Horticulture and Natural Resources, Kansas State University, 1602 Throckmorton Hall, 1712 Claflin Road, Manhattan, KS 66506, USA. 3U.S. Geological Survey, Kansas Cooperative Fish and Wildlife Research Unit, Division of Biology, Kansas State University, 205 Leasure Hall, 1128 N. 17th Street, Manhattan, KS 66506, USA. 4Department of Fisheries and Wildlife, Oregon State University, 104 Nash Hall, 2820 S.W. Campus Way, Corvallis, OR 97331, USA. 5Kansas Department of Wildlife, Parks, and Tourism, 1830 Merchant Street, Emporia, KS 66801, USA. *Corresponding author: msirch@uvm.edu
Praire Naturalist, Volume 54 (2022):11–23
Abstract
Lack of fire in contemporary grasslands has contributed to the invasion of woody plants that can survive low-intensity fire upon maturity, but knowledge of the effects of megafire (>100,000 ac [40,000 ha]) on mortality of woody species within grasslands is limited. We used remote sensing and ground surveys to estimate tree canopy cover change and rates of top-kill and mortality of woody species in the mixed-grass prairie following the Starbuck megafire. After the megafire, we detected 63% tree canopy cover loss, 17 ± 4% (mean ± SD) tree death, and 56 ± 5% tree top-kill with resprouting. We conclude that further postfire management (e.g., targeted herbicide, mechanical removal) may be required to limit woody encroachment following megafire.
Download Full-text pdf
11
1Kansas Cooperative Fish and Wildlife Research Unit, Division of Biology, Kansas State University, 205
Leasure Hall, 1128 N. 17th Street, Manhattan, KS 66506, USA. 2Department of Horticulture and Natural
Resources, Kansas State University, 1602 Throckmorton Hall, 1712 Claflin Road, Manhattan, KS 66506,
USA. 3U.S. Geological Survey, Kansas Cooperative Fish and Wildlife Research Unit, Division of Biology,
Kansas State University, 205 Leasure Hall, 1128 N. 17th Street, Manhattan, KS 66506, USA. 4Department
of Fisheries and Wildlife, Oregon State University, 104 Nash Hall, 2820 S.W. Campus Way, Corvallis, OR
97331, USA. 5Kansas Department of Wildlife, Parks, and Tourism, 1830 Merchant Street, Emporia, KS
66801, USA. *Corresponding author: msirch@uvm.edu
Associate Editor: Sean Jenkins, Department of Biological Sciences, Western Illinois University.
2022 PRAIRIE NATURALIST 54:11–23
Effects of Megafire on Woody Species in the
Mixed-Grass Prairie
Matthias W. Sirch1*, Daniel S. Sullins2, Nicholas J. Parker2, David A. Haukos3,
John D. Kraft1, Christian A. Hagen4, and Kent A. Fricke5
Abstract - Lack of fire in contemporary grasslands has contributed to the invasion of woody plants that can
survive low-intensity fire upon maturity, but knowledge of the effects of megafire (>100,000 ac [40,000 ha]) on
mortality of woody species within grasslands is limited. We used remote sensing and ground surveys to estimate
tree canopy cover change and rates of top-kill and mortality of woody species in the mixed-grass prairie following
the Starbuck megafire. After the megafire, we detected 63% tree canopy cover loss, 17 ± 4% (mean ±
SD) tree death, and 56 ± 5% tree top-kill with resprouting. We conclude that further postfire management (e.g.,
targeted herbicide, mechanical removal) may be required to limit woody encroachment following megafire.
Introduction
Fire within the mixed-grass prairie has historically been a key ecological driver in maintaining
treeless landscapes (Askins et al. 2007, Gleason 1913, Keeley and Rundel 2005). However,
land management in the form of fire suppression and intensive livestock grazing has encouraged
encroachment of woody species onto grasslands (Briggs et al. 2005, DeSantis et al. 2011, Engle
et al. 2008, Van Auken 2000). Such alterations, alongside land conversion for row-crop agriculture
and energy development, have resulted in a mixed-grass prairie ecoregion with grasslands
that are fragmented and invaded by woody and non-native species (Dodson and Fiedler 2006,
Keeley 2006, Samson et al. 2004). As woody biomass increases, the additional fuels can promote
megafires, a term typically defined as a fire burning >100,000 ac (40,000 ha) (Bracmort
2013, Brown 2018, Lindley et al. 2019).
Today, the Great Plains region is recognized as one of six global hotspots for high grassland
wildfire risk, and climate change models predict future increases of very large wildfires throughout
the region (Barbero et al. 2015, Cao et al. 2015, Donovan et al. 2017). In support of these predictions,
the two largest wildfires recorded in Kansas history occurred in 2016 and 2017 within the
southern mixed-grass prairie of the Great Plains. The 2017 fire, known as the Starbuck Fire, burned
approximately 662,737 ac (268,200 ha) from March 6th to 10th (Lindley et al. 2019).
The Starbuck Fire burned 41,374 ac (16,744 ha, 6% of burned area) at moderate severity
and 526,890 ac (213,225 ha, 80%) at low severity, with the remaining 91,737 ac (37,125 ha,
14%) classified as unburned or charred (Monitoring Trends in Burn Severity 2020) by using the
differenced Normalized Burn Ratio (dNBR) index. In contrast to forests, grasslands have minimal
fuel loads and grassland fires spread and burn through an area rapidly, accounting for the
Prairie Naturalist
M W. Sirch, D. S. Sullins, N. J. Parker, D. A. Haukos, J. D. Kraft, C. A. Hagen, K. A. Fricke
2022 54:11–23
12
low severity as the fire removes less vegetation and moisture from the soil than slower moving
forest fires (Dillon et al. 2020, Keeley 2009, Miller et al. 2009). Accordingly, even areas burned
at low severity experience a near-complete removal of vegetation. This low severity estimate for
the Starbuck Fire is similar to all other recent large grassland fires (eight wildfires >50,000 ac
[20,000 ha], 2016–2018) in the mixed-grass prairie (Monitoring Trends in Burn Severity 2020).
The Starbuck Fire extensively burned portions of one of the largest remaining tracts of
intact grasslands in the region, which provides habitat for several vulnerable grassland-dependent
wildlife species, including Tympanuchus pallidicinctus Ridgway (Lesser Prairie-chicken;
Haukos and Boal 2016, Rosenberg et al. 2019). As Lesser Prairie-chickens are sensitive to tall
features (>3.28 ft [1 m]) on the landscape (Boggie et al. 2017, Falkowski et al. 2017, Lautenbach
et al. 2017, Plumb et al. 2019), woody encroachment has become a regionally important threat to
maintaining grassland habitat for wildlife. Prescribed burning is frequently used to limit woody
encroachment, which, when incorporated with mechanical tree removal, can control woody
encroachment and produce habitat for grassland-dependent species in the mixed-grass prairie
(Fuhlendorf et al. 2017, Lautenbach 2017, Ratajczak et al. 2014).
Two woody species that have substantially encroached onto the Great Plains include the native
Juniperus virginiana L. (Eastern Redcedar) and non-native Tamarix ramosissima Ledeb.
(Saltcedar). Encroachment by Eastern Redcedar can reduce grassland biodiversity, lower livestock
productivity, increase carbon volatilization, and reduce wildfire suppression potential (Ratajczak
et al. 2012, Twidwell et al. 2013). Several studies of Eastern Redcedar and other Juniperus species
have demonstrated high postfire mortality rates that decrease as trees mature or surrounding
vegetation is cut or grazed (Briggs et al. 2002, Noel and Fowler 2007, Ratajczak et al. 2014, Weir
and Engle 2017). For Saltcedar, prescribed burning mortality rarely exceeds 30% as it vigorously
sprouts from roots following most disturbances (Delwiche 2009, Hohlt et al. 2002, Racher et al.
2003). Further, Saltcedar serves as a fuel ladder that propels flaming embers onto surrounding
vegetation beyond the fire perimeter, making prescribed burning challenging (Delwiche 2009).
Because controlled wildfire experiments are not easily replicated, the effects of wildfire
on vegetation should be evaluated when possible. Although ecological effects of prescribed
burning have been studied extensively, vegetation responses to prescribed burning cannot be
directly transcribed to vegetation exposed to more intensive wildfires such as megafires (Arkle
and Pilliod 2010, Eisenberg et al. 2019). Most notably, prescribed burns are conducted in predictable
low-wind and moderate-high humidity conditions, even during extreme and large-scale
experiments (Bidwell et al. 2018, Dey and Hartman 2004, Weir 2014). As landscape-scale and
intensive grassland wildfires in the Great Plains become more frequent (Donovan et al. 2017,
Polley et al. 2013, Pyne 2017), knowledge of how megafires influence woody species in the
highly altered mixed-grass prairie remains limited.
To better understand the effects of megafires within the mixed-grass prairie, our objectives
were to (1) estimate tree canopy change from ground surveys and remote-sensing techniques,
and (2) evaluate woody species conditions including rates of dead and top-killed plants. We
hypothesized that mortality of woody vegetation following the Starbuck Fire would be high but
vary by species, and canopy cover would be substantially reduced.
Materials and Methods
Study Area
We examined the effects of the 2017 Starbuck Fire on the mixed-grass prairie of Clark County,
Kansas, USA (Fig. 1). The climate was characterized by a warm growing season (mid-April to mid-
October) followed by months of little moisture, with precipitation averaging 22.25 in (56.49 cm)
annually between 1895 and 2018 (NOAA 2020a). Wildfires typically occur in March and April as
Prairie Naturalist
M W. Sirch, D. S. Sullins, N. J. Parker, D. A. Haukos, J. D. Kraft, C. A. Hagen, K. A. Fricke
2022 54:11–23
13
vegetation has senesced after the dry period of December–February, which on average receives 0.66
in (1.67 cm) of precipitation per month (Lindley et al. 2019, NOAA 2020a). Wind speed averages
are also highest during March and April as air pressure fluctuates due to a convergence of warm air
from the Gulf of Mexico and cold air from the Rocky Mountains and the Arctic (Jones and Cushman
2004). Fire return intervals are thought to range 1 to 10 years with an average fire return interval of
5 to 10 years (Fryer and Luensmann 2012, Joern and Keeler 1995).
The warm and dry conditions preceding the 6–10 March 2017 Starbuck Fire intensified effects
of the megafire (Lindley et al. 2019). October, November, and December 2016 were particularly
dry, receiving less than 1.41 in (3.58 cm) of precipitation cumulatively, 0.63 in (1.6 cm) below
historical average (1895–2016) (NOAA 2020a). February 2017 accumulated only 0.07 in (0.18
cm) of precipitation with mean temperatures 9° F (5° C) above historical average (1895–2016)
(NOAA 2020a), further desiccating vegetation. Fire ignited by a downed power line in western
Oklahoma quickly spread northward on 6 March 2017. Only five hours after ignition, winds of 35
knots (40 mph [65 km/h]), with gusts of 50 knots (57 mph [92 km/h]), pushed flames the height of
utility poles (~30 ft [9 m]) from the southwest before rapidly and repeatedly changing directions
(Frazier 2018, NOAA 2020b). The Starbuck Fire burned most actively and expanded in size over
a two-day period (March 6–7, 2017) and has been described as a short-duration evolution type fire
(Lindley et al. 2019). In total, the fire burned an estimated 662,737 ac (268,200 ha).
We conducted ground surveys to evaluate the effects of the fire on 38,500 ac (15,580 ha)
of accessible private land within Clark County, Kansas (Fig. 1). The mosaic of land cover types
included native grass pastures, Conservation Reserve Program (CRP) fields, alkali flats, sand
dunes, and row-crop agriculture. Trees, primarily Populus deltoides W. Bartram ex Marshall
(Eastern Cottonwood), Sapindus saponaria L. (Western Soapberry), Gleditsia triacanthos L.
(Honey Locust), and Maclura pomifera (Raf.) C.K. Schneid (Osage Orange), generally grow in
windbreaks along agricultural field edges. Several woody species, including Eastern Redcedar,
Figure 1. Plot network established to survey postfire tree conditions after the 2017 Starbuck Fire across
accessible property in Clark County, Kansas, and Beaver County, Oklahoma. Ground surveys were
conducted between 28 August and 1 October 2018, on woodland (n = 98) and Saltcedar plots (n = 60).
The differenced Normalized Burn Ratio (dNBR) severity index of the Starbuck Fire was calculated by
the Monitoring Trends in Burn Severity program. Tree canopy cover loss was estimated from change in
June 2015 and July 2017 National Agricultural Imagery Program (NAIP) imagery.
Prairie Naturalist
M W. Sirch, D. S. Sullins, N. J. Parker, D. A. Haukos, J. D. Kraft, C. A. Hagen, K. A. Fricke
2022 54:11–23
14
Ulmus L. spp. (Elm spp.), Elaeagnus angustifolia L. (Russian Olive), and Salix nigra Marshall
(Black Willow) have encroached into the grasslands. Saltcedar has established in dense stands
along riverine lowlands of the Cimarron River.
Change in Canopy Cover
To remotely estimate tree canopy cover, we examined defined woodland areas using the
2015 USDA Forest Service Tree Canopy layer, compiled by Paull et al. (2017), throughout the
study area. We manually refined the woodland canopy layer across accessible private land using
1-meter resolution National Agricultural Imagery Program (NAIP) imagery for June 2015 and
July 2017 to identify tree canopy cover both before and after the wildfire. Tree canopy was distinguished
from surrounding vegetation by a distinct shadow at a 1:1000 scale. Pre- and postfire
tree canopy layers were manually drawn using ArcGIS 10.6 (ESRI 2018).
Field Collected Data
To ground-truth our remotely-sensed tree canopy estimates and estimate woody species mortality
on the ground, we randomly generated 100 non-overlapping 65.62 ft (20 m) diameter woodland
ground survey plots within the 2015 pre-fire woodland canopy layer. Two woodland plots
were later excluded because of documented aerial application of herbicide in the area. We then
used the remotely-sensed postfire 2017 canopy values to compare with 2018 field observations.
We evaluated the relationship of remotely-sensed and woodland ground survey canopy estimates
using a generalized linear regression in Program R 3.6.1 (R Development Core Team 2019).
Our woodland plots mostly focused on wooded uplands and hedgerows interspersed
throughout the predominantly grassland study area. To better evaluate Saltcedar mortality following
the Starbuck Fire, we established an additional 60 plots (Saltcedar plots) centered in
riparian areas largely invaded by Saltcedar. The areas were manually delineated in ArcGIS 10.6
using the 2015 NAIP imagery to identify Saltcedar stands by their characteristic clustering. We
randomly generated 30 plots in both burned and unburned areas within Saltcedar clusters. Because
there were no accessible unburned Saltcedar clusters in Clark County, we established the
unburned Saltcedar plots within accessible private property in Beaver County, Oklahoma, ~7 mi
(11 km) from the closest burned plot (Fig. 1).
For each 65.62 ft (20 m) diameter woodland and Saltcedar plot, we estimated composition
of woody species and percentage of intact, dead, and top-killed plants. Top-kill was defined as
appearing dead except for new resprout branches growing from the living root system, whereas
intact individuals had living canopy branches. We tallied plants as an individual if the diameter
at breast height (DBH) was >1 in (2.54 cm) and branches appeared to have emerged from the
same root system. For every intact, dead, and top-killed woody species, we measured height of
the plant and height of the tallest resprouting branch. In each plot, we also measured largest DBH
of each species. We ocularly estimated heights between 10 ft (3 m) and 30 ft (10 m) and used a
clinometer for branches taller than 30 ft (10 m). For each woodland plot, we estimated canopy
cover ocularly in groups of 0, 1–5, 5–25, 25–50, 50–75, 75–95, and 95–100%. We estimated
canopy cover ocularly rather than with a densiometer because of thick understory (<16 ft [5 m]
tall; Korhonen et al. 2006). We conducted ground surveys a year and a half after the wildfire,
between 28 August and 1 October 2018.
Estimation of Woody Species Mortality
To evaluate megafire effects on each woody species, we estimated the proportion that
were dead and top-killed within burned plots. We estimated proportions of mortality and
top-kill among all plots, SD of proportions as where p is proportion, q is (1 – p), and n is
Prairie Naturalist
M W. Sirch, D. S. Sullins, N. J. Parker, D. A. Haukos, J. D. Kraft, C. A. Hagen, K. A. Fricke
2022 54:11–23
15
number of samples, and adjusted Wald 95% confidence intervals following Zar (2010). If p
= 0 or 1, we used confidence intervals derived from nth root of α = 0.025 following methods
in Zar (2010).
Analysis of Woody Species Mortality
To evaluate if tree mortality and top-kill rate varied by species, we logit transformed
percent dead and top-kill estimates as our response variables for each ground survey plot and
then performed a multivariate analysis of variance (MANOVA) with burned and unburned
treatments as independent variables (Warton and Hui 2011). Although MANOVA can handle
correlated dependent variables, we examined correlation between mortality and top-kill rates
using a Pearson’s product-moment correlation test. We then performed an analysis of variance
(ANOVA) to evaluate influence of fire on logit transformed dead and resprout rates separately.
Finally, we performed Welch’s t-tests to examine woody species height and DBH difference
between burned and unburned plots. All analyses were conducted using Program R 3.6.1 (R
Development Core Team 2019).
Results
Change in Canopy Cover
Our remotely-sensed analysis indicated a 63% woodland canopy cover loss within the
burned portion of the study area, decreasing from 126.5 ac (51.2 ha) pre-fire (0.33% of study
area) to 46.95 ac (19.0 ha, 0.12%) postfire. Within woodland plots, remotely-sensed canopy
cover decreased from 61.8 ± 31.2% to 18.9 ± 28.8% after the wildfire. This woodland canopy
estimate using remotely-sensed data was comparable to canopy cover estimated at postfire
ground-truth plots (11.0 ± 20.4%), with remotely-sensed canopy slightly greater than ground
canopy within woodland plots (Yremotely sensed = 1.01 * Xground estimate + 7.82; r2 = 0.51, t153.2 =
2.1, P = 0.04). For Saltcedar plots, ground-measured postfire canopy cover was 0.6 ± 3% of the
surveyed area.
Species Composition
Each woodland plot contained 1.7 ± 0.9 tree species on average. Woody species within
woodland plots were predominantly Honey Locust (26 ± 4%), Western Soapberry (23 ± 4%),
Osage Orange (13 ± 3%), Elm spp. (9 ± 3%), Saltcedar (6 ± 2%), Eastern Cottonwood (5 ±
2%), and Eastern Redcedar (4 ± 2%). Artemisia filifolia Torr. (Sand Sagebrush) (5 ± 13%) and
Prunus angustifolia Marshall (Sand Plum) (5 ± 12%) were the predominant shrub species.
Within Saltcedar plots, Saltcedar accounted for 97 ± 2% of woody species (Table 1).
Estimation of Woody Species Mortality
We collected data on woody species mortality and condition after the Starbuck Fire. Overall,
17 ± 4% of trees were dead within burned woodland plots while 56 ± 5% were top-killed
and resprouted from roots (Table 2). Percentages of dead individuals were greatest for Eastern
Redcedar (100%) and Eastern Cottonwood (86 ± 3%), and lowest for Russian Olive and Black
Willow (0%) (Table 2). Of 117 burned ground survey plots, 116 plots (99%) had a tallest branch
at >3.28 ft (1 m) tall. Tallest branches in burned ground survey plots averaged 29.36 ± 14.60
ft (8.95 ± 4.45 m).
Percentage of dead and top-killed tree species in each plot were negatively correlated,
which reflects that if trees were not killed by fire they typically were top-killed (r = -0.62, t391=-
15.41, P < 0.001). Combined mortality and top-kill rates varied among burned and unburned
treatments (Wilk’s λ = 0.53, F2,357 = 159.5, P < 0.001) and by species (Wilk’s λ = 0.48, F26,714 =
Prairie Naturalist
M W. Sirch, D. S. Sullins, N. J. Parker, D. A. Haukos, J. D. Kraft, C. A. Hagen, K. A. Fricke
2022 54:11–23
16
12.3, P < 0.001). The interactive effect of the burn on each species was also significant, indicating
that mortality and top-kill rates from the megafire were species-specific (Wilk’s λ = 0.87,
F18,714 = 2.8, P < 0.001). Black Willow, Russian Olive, Honey Locust, Western Soapberry, and
Osage Orange all showed high rates of top-kill, with branches resprouting from the tree base
(Table 2). Among shrub species, Sand Sagebrush exhibited the greatest mortality, followed by
Sand Plum and Rhus aromatica Aiton (Fragrant Sumac).
Saltcedar Mortality and Top-kill
Our comparison among burned and unburned treatments revealed that Saltcedar was
largely top-killed by the megafire and resprouted from the base. In all burned ground survey
plots, 33 ± 9% of Saltcedar were dead, 65 ± 9% were top-killed, and 2 ± 4% remained living
with intact canopy cover (Table 2). Percentage of dead and top-kill Saltcedar varied among
burned and unburned plots (Wilk’s λ = 0.20, F2,149 = 292.6, P <0.001). However, this difference
was mostly due to the difference in top-kill rates among burned and unburned treatments
(F1,150 = 88.8, P < 0.001). In unburned plots, Saltcedar were 21 ± 8% dead, 1 ± 3% top-killed,
and 78 ± 8% intact.
Tallest Saltcedar branches in burned plots were on average 12.40 ± 2.46 ft (3.78 ± 0.75
m) tall, similar to 13.19 ± 2.20 ft (4.02 ± 0.67 m) unburned Saltcedar branches (t58.693
= -1.31, P = 0.19). We observed 1 dead Saltcedar >16.4 ft (5 m) tall and 4 dead Saltcedar
>13.1 ft (4 m) tall, indicating the megafire likely killed large Saltcedar trees. A year and a
half after the Starbuck Fire, resprouting branches of top-killed Saltcedar grew to 6.40 ± 1.74
ft (1.95 ± 0.53 m) tall.
Woody species
Trees Honey Locust 0.26 0.04 83 149 0 NA 0 NA
W. Soapberry 0.23 0.04 76 247 0 NA 0 NA
Osage Orange 0.13 0.03 42 91 <0.01 <0.01 3 21
Elm Spp. 0.09 0.03 31 71 <0.01 <0.01 7 25
Saltcedar 0.06 0.02 21 103 0.97 0.02 1514 1282
E. Cottonwood 0.05 0.02 15 35 0 NA 0 NA
E. Redcedar 0.04 0.02 12 52 <0.01 <0.01 6 15
Mulberry 0.02 0.01 6 26 0 NA 0 NA
Black Locust 0.02 0.01 6 30 0 NA 0 NA
Black Willow <0.01 <0.01 <1 3 0 NA 0 NA
Ash Spp. <0.01 <0.01 1 10 0 NA 0 NA
Russian Olive 0 NA 0 NA <0.01 <0.01 4 19
Overall 0.9 0.03 293 272 0.98 0.03 1534 1278
Shrubs Sand Sagebrush 0.05 0.02 16 86 0.01 0.02 23 106
Sand Plum 0.05 0.02 15 108 0 NA 0 NA
Fragrant Sumac <0.01 <0.01 <1 3 <0.01 <0.01 2 9
Overall 0.1 0.02 31 148 0.02 0.02 25 111
Woodland plots Saltcedar plots
Plants/ha Plants/ha
x̅ SD x̅ SD x̅ SD x̅ SD
Table 1. Overall mean woody species composition and standard deviation within woodland (n = 98) and
Saltcedar plots (n = 60) surveyed in Clark County, Kansas, and Beaver County, Oklahoma, USA between
28 August and 1 October 2018.
Prairie Naturalist
M W. Sirch, D. S. Sullins, N. J. Parker, D. A. Haukos, J. D. Kraft, C. A. Hagen, K. A. Fricke
2022 54:11–23
17
Table 2. Tree and shrub species counts and proportion of dead and top-killed following the Starbuck Fire within all burned ground survey plots (n = 117) in
Clark County, Kansas, USA between 28 August and 1 October 2018. Plots were established where trees were remotely detected in 2015.
Dead (proportion) Top-kill (proportion)
Trees E. Redcedar 35 9 1 0 0.91–1.00 0 0 0.00–0.09
E. Cottonwood 42 21 0.86 0.03 0.72–0.94 0.05 0.02 0.00–0.17
Ash Spp. 3 1 0.67 0.04 0.20–0.94 0.33 0.04 0.06–0.80
Mulberry 10 4 0.4 0.05 0.17–0.69 0.3 0.04 0.10–0.61
Saltcedar 901 37 0.33 0.09 0.25–0.43 0.65 0.09 0.55–0.74
Elm Spp. 97 28 0.22 0.04 0.15–0.31 0.42 0.05 0.33–0.52
Honey Locust 219 40 0.1 0.03 0.06–0.14 0.75 0.04 0.69–0.81
Osage Orange 99 24 0.08 0.03 0.04–0.15 0.53 0.05 0.43–0.62
Black Locust 18 5 0.06 0.02 0.00–0.28 0.78 0.04 0.54–0.92
W. Soapberry 225 16 0.02 0.01 0.01–0.05 0.54 0.05 0.41–0.54
Black Willow 1 1 0 NA 0.00–0.78 1 NA 0.22–1.00
Russian Olive 3 2 0 0 0.00–0.53 1 0 0.47–1.00
Overall Woodland Plots 872 87 0.17 0.04 0.12–0.29 0.56 0.05 0.54–0.74
Overall Saltcedar Plots 873 30 0.35 0.09 0.26–0.44 0.63 0.09 0.53–0.72
Overall 1796 117 0.26 0.04 0.16–0.31 0.6 0.05 0.42–0.60
Shrubs Sand Sagebrush 92 9 0.08 <0.01 0.03–0.15 0.92 0.03 0.85–0.97
Sand Plum 47 2 0.02 <0.01 0.00–0.12 0.98 0.01 0.88–1.00
Fragrant Sumac 3 2 0 0 0.00–0.53 1 0 0.47–1.00
Woody species No. total No. plots present x̅ SD 95% CI x̅ SD 95% CI
Prairie Naturalist
M W. Sirch, D. S. Sullins, N. J. Parker, D. A. Haukos, J. D. Kraft, C. A. Hagen, K. A. Fricke
2022 54:11–23
18
Discussion
Variation in postfire mortality and top-kill rates by species in our study highlights the complex
nature of woodland response to disturbance, even in areas with limited tree diversity such
as our study site in the mixed-grass prairie of western Kansas. Our study contributes to a small
but growing body of evidence on the impacts of megafires on grassland plant communities in
the Great Plains (Donovan et al. 2020, Rideout-Hanzak et al. 2011), and specifically advances
knowledge on the impact to woody species interspersed within grassland-dominated landscapes.
Results indicated that megafire in mixed-grass prairie initially reduced canopy cover, but many
burned deciduous trees resprouted from their root systems. For example, complete mortality
of Saltcedar was not substantially greater in burned areas relative to unburned areas. However,
Saltcedar in a burned area experienced a 65% top-kill rate that removed nearly all riverine
canopy cover. In contrast to the high prevalence of resprouting from top-killed Saltcedar and
other deciduous species, we documented substantial mortality among Eastern Cottonwoods and
complete mortality for the invasive Eastern Redcedar.
Our results suggest that megafires historically restricted woody encroachment into grasslands
in the Great Plains by eliminating mature stands of Juniperus and Populus spp. Eastern
Cottonwoods that died following the Starbuck Fire were fairly tall, with at least 6 individuals
>60 ft (18 m) tall and >31.5 in (80 cm) DBH. Resprouting was also limited for many Eastern
Cottonwood across the study area. In contrast to our findings, Wonkka et al. (2018) observed
high Eastern Cottonwood regeneration rates following an early May wildfire ~105 m (170 km)
west of our study area. Differences in the size or timing of fires, as well as location of trees
among hedgerows in our study rather than in riparian areas, may have influenced conflicting
regeneration rates between the two studies. In addition, our reduced regeneration may have been
caused by an interaction of browsing on resprouting branches by cattle or deer and the effect of
the fire (Glinski 1977, Stromberg 1997).
Our high postfire mortality estimates for Eastern Cottonwood and Eastern Redcedar were
substantially greater than estimates historically observed in the region’s woodlands. Percentages
of dead Populus and Juniperus spp. across Kansas in typical unburned woodlands have been estimated
by the USDA Forest Inventory and Analysis (FIA) Program as 19 ± 3% and 2 ± 1%, respectively
(>5 in [12.7 cm] DBH, 2006–2016). The FIA standing dead baseline for Populus spp.
was the greatest of all examined species, though still well below the 86% we recorded following
the megafire (Table 2). This suggests substantial mortality of Eastern Cottonwood despite the
thick insulating bark for which the species is known (VanderWeide and Hartnett 2011). For Eastern
Redcedar, we documented complete mortality of all trees following the megafire, even trees
>20 ft (6 m) tall and with DBH >10 in (25 cm). An Eastern Redcedar of this size would more
likely survive a smaller and less intense prescribed fire (Briggs et al. 2002, Nippert et al. 2021,
Owensby et al. 1973), but the crown damage of the megafire’s larger flames likely killed the
mature juniper trees (Weir and Engle 2017). Eastern Redcedar lacks the adventitious buds at the
root collar that allow other fire-adapted woody species to resprout (Glasgow and Bidwell 2009).
Although our reported mortality rates suggest megafires may have restricted encroachment
of native Eastern Redcedar and Eastern Cottonwood into mixed grass prairie, mortality of the
exotic invasive Saltcedar was substantially lower. The Saltcedar mortality rate of 33% from this
study coincides with rates of prescribed burn studies across the region (Delwiche 2009, Hohlt et
al. 2002, Racher et al. 2003). However, Saltcedar mortality was not substantially different between
burned and unburned areas. Our comparison among burned and unburned areas may have been
confounded by Diorhabda spp. Desbrochers (Tamarisk Beetle). The relatively high Saltcedar mortality
rate in unburned areas was likely caused by Tamarisk Beetles, which have been introduced
across the Southwest since 2001 as a Saltcedar defoliator biocontrol (Bean and Dudley 2018,
Prairie Naturalist
M W. Sirch, D. S. Sullins, N. J. Parker, D. A. Haukos, J. D. Kraft, C. A. Hagen, K. A. Fricke
2022 54:11–23
19
Hatten 2016) and has recently expanded into portions of the study area. Although the Saltcedar
mortality rate was low, we documented near-complete removal of canopy from riverine lowland
Saltcedar stands. Additionally, our estimates suggest that 65% of Saltcedar were top-killed and
resprouted, with new sprouts growing to nearly 6.5 ft (2 m) tall over a year and a half after the fire.
This rapid growth rate corresponds with previous research noting faster growth rates for
recently disturbed Saltcedar (Goldsmith and Smart 1982). We recognize that wildfire alone has
limited potential for returning established Saltcedar stands to grassland habitat, particularly due
to this resprouting capacity and the species’ ability to disperse millions of seeds annually (Bond
and Midgely 2003, Di Tomaso 1998, Stevens 2002). Similar to patterns of Saltcedar, we documented
high resprouting potential in several other deciduous taxa, including Elm spp., Honey
Locust, Osage Orange, and Western Soapberry (Bond and Midgely 2003). Our results suggest
that future megafires, if not selecting for herbaceous cover, will select for native or invasive
non-native woody species with strong resprouting potential, similar to postfire woody vegetation
shifts in Arizona following decades of fire suppression (Taylor et al. 2021).
In addition to better understanding direct responses of both exotic and native invasive woody
species, the accurate prediction of future woody species response to megafire will need to account
for interactions with other ecological drivers including climate (Askins et al. 2007). As
the mixed-grass prairie experiences warmer and drier conditions, greater variability of weather
linked to climate change could result in more intense wet seasons followed by intense drought
during dormant seasons. Such a scenario would promote the growth and drying of grasses as fuel
for fire (Courtwright 2011). When combined with increased woody encroachment throughout the
region, megafires may become more prominent and challenging to suppress (Lindley et al. 2019).
Megafires like the Starbuck Fire have the potential to reduce the abundance of trees on the landscape,
creating habitat for grassland wildlife species. However, once undesirable trees become
established, it will require substantial disturbance and effort for the woody species to be removed.
Management Implications
Even following a megafire, resprouting Saltcedar and other tree species were prevalent.
Mechanical removal of root systems or multiple years of herbicide may be required to eliminate
Saltcedar from an area (Duncan and McDaniel 2009, Fick and Geyer 2010, McDaniel and Taylor
2003). In areas suitable for Saltcedar, aggressive efforts to prevent establishment of Saltcedar
and other resprouting species may be more cost effective. In contrast, large fires and megafires
may present an opportunity to reclaim grasslands previously invaded by Eastern Redcedar.
Although Eastern Redcedar will not resprout, dead Eastern Redcedar may continue to stand
for many years, functionally limiting habitat availability for grassland-obligate species like the
Lesser Prairie-chicken (Fuhlendorf et al. 2017, Lautenbach et al. 2017, Oberle et al. 2018).
Acknowledgments
We thank Corrie Desilets who helped collect field data and all private landowners for allowing access.
For financial and logistical support, we thank Kansas State University, Kansas Department of Wildlife,
Parks, and Tourism, United States Department of Agriculture (USDA) Natural Resources Conservation
Service, Pheasants Forever, and Lesser Prairie-Chicken Initiative. The research which is the subject of this
manuscript has been financed, in part, with federal funds from the Fish and Wildlife Service, a division of
the United States Department of Interior, and administered by the Kansas Department of Wildlife, Parks,
and Tourism. The contents and opinions, however, do not necessarily reflect the views or policies of the
United States Department of Interior or the Kansas Department of Wildlife, Parks and Tourism. Any use
of trade, firm, or product names is for descriptive purposes only and does not imply endorsement by the
U.S. Government.
Prairie Naturalist
M W. Sirch, D. S. Sullins, N. J. Parker, D. A. Haukos, J. D. Kraft, C. A. Hagen, K. A. Fricke
2022 54:11–23
20
Literature Cited
Arkle, R.S., and D.S. Pilliod. 2010. Prescribed fires as ecological surrogates for wildfires: A stream and
riparian perspective. Forest Ecology and Management 259:893–903.
Askins, R.A., F. Chávez-Ramírez, B.C. Dale, C.A. Haas, J.R. Herkert, F.L. Knopf, and P.D. Vickery.
2007. Conservation of grassland birds in North America: Understanding ecological processes in different
regions. Ornithological Monographs 64:1–46.
Barbero, R.J., T. Abatzoglou, N.K. Larkin, C.A. Kolden, and B. Stocks. 2015. Climate change presents increased
potential for very large fires in the contiguous United States. International Journal of Wildland
Fire 24:892–899.
Bean, D., and T. Dudley. 2018. A synoptic review of Tamarix biocontrol in North America: Tracking success
in the midst of controversy. BioControl 63:361–376.
Bidwell, T.G., J.R. Weir, J.D. Carlson, R.E. Masters, and D.M. Engle. 2018. Fire prescriptions for restoration
and maintenance of native plant communities. Oklahoma Cooperative Extension Service Report
F-2878. Stillwater, OK, USA. 8 pp.
Boggie, M.A., C.R. Strong, D. Lusk, S.A. Carleton, W.R. Gould, R.L. Howard, C. Nichols, M.J.
Falkowski, and C.A. Hagen. 2017. Impacts of mesquite distribution on seasonal space use of Lesser
Prairie-chickens. Rangeland Ecology and Management 70:68–77.
Bond, W.J., and J.J. Midgley. 2003. The evolutionary ecology of sprouting in woody plants. International
Journal of Plant Sciences 164:S103–S114.
Bracmort, K. 2013. Wildfire fuels and fuel reduction. Congressional Research Service Report R40811.
15 pp. Washington, D.C., United States.
Briggs, J.M., G. Hoch, and L. Johnson. 2002. Assessing the rate, mechanisms, and consequences of the
conversion of tallgrass prairie to Juniperus virginiana forest. Ecosystems 5:578–586.
Briggs, J.M., A.K. Knapp, J.M. Blair, J.L. Heisler, G.A. Hoch, M.S. Lett, and J.K. Mccarron. 2005. An
ecosystem in transition: Causes and consequences of the conversion of mesic grassland to shrubland.
BioScience 55:243–254.
Brown, H. 2018. Fire control and the 2015 Canyon Creek Complex fire. Fire Management Today
76:36–40.
Cao, X., Y. Meng, and J. Chen. 2015. Mapping grassland wildfire risk of the world. Pp. 277–283, In P. Shi
and R. Kasperson (Eds.). World Atlas of Natural Disaster Risk. Springer-Verlag, Berlin, Heidelberg,
Germany. 368 pp.
Courtwright, J. 2011. Prairie Fire: A Great Plains History. University Press of Kansas, Lawrence, KS,
USA. 288 pp.
Delwiche, J. 2009. Saltcedar: Is burning an option? Joint Fire Science Program Briefs Issue 50. Boise, ID,
USA. 7 pp.
DeSantis, R.D., S.W. Hallgren, and D.W. Stahle. 2011. Drought and fire suppression lead to rapid forest
composition change in a forest-prairie ecotone. Forest Ecology and Management 261:1833–1840.
Dey, D.C., and G. Hartman. 2004. Effects of different large-scale prescribed burning regimes of advanced
production in Missouri Ozarks. Pp. 168–176, In M.A. Spetich (Ed.). Proceedings of the Upland Oak
Ecology Symposium: History, current conditions, and sustainability. US Department of Agriculture,
Forest Service, Southern Research Station General Technical Report SRS-73. Asheville, NC, USA.
311 pp.
Di Tomaso, J.M. 1998. Impact, biology, and ecology of Saltcedar (Tamarix spp.) in the southwestern
United States. Weed Technology 12:326–336.
Dillon, G.K., M.H. Panunto, B. Davis, P. Morgan, D.S. Birch, and W.M. Jolly. 2020. Development of
a severe fire potential map for the contiguous United States. US Department of Agriculture, Forest
Service, Rocky Mountain Research Station General Technical Report RMRS-GTR-415. Fort Collins,
CO, USA. 107 pp.
Dodson, E.K., and C.E. Fiedler. 2006. Impacts of restoration treatments on alien plant invasion in Pinus
ponderosa forests, Montana, USA. Journal of Applied Ecology 43:887–897.
Donovan, V.M., C.L. Wonkka, and D. Twidwell. 2017. Surging wildfire activity in a grassland biome.
Geophysical Research Letters 44:5986–5993.
Prairie Naturalist
M W. Sirch, D. S. Sullins, N. J. Parker, D. A. Haukos, J. D. Kraft, C. A. Hagen, K. A. Fricke
2022 54:11–23
21
Donovan V.M., D. Twidwell, D.R. Uden, T. Tadesse, B.D. Wardlow, C.H. Bielski, M.O. Jones, B.W.
Allred, D.E. Naugle, and C.R. Allen. 2020. Resilience to large, “catastrophic” wildfire in North
America’s grassland biome. Earth’s Future 8, e2020EF001487.
Duncan, K., and K. McDaniel. 2009. Chemical weed and brush control for New Mexico rangelands. New
Mexico State University Cooperative Extension Service Circular 579. Las Cruces, NM, USA. 20 pp.
Eisenberg, C., C.L. Anderson, A. Collingwood, R. Sissons, C.J. Dunn, G.W. Meigs, D.E. Hibbs, S.
Murphy, S.D. Kuiper, J. SpearChief-Morris, L. Little Bear, B. Johnston, and C.B. Edson. 2019. Out
of the ashes: Ecological resilience to extreme wildfire, prescribed burns, and Indigenous burning in
ecosystems. Frontiers in Ecology and Evolution 7:436.
Engle, D.M., B.R. Coppedge, and S.D. Fuhlendorf. 2008. From the dust bowl to the green glacier: Human
activity and environmental change in Great Plains grasslands. Pp. 253–271, In O.W. Van Auken (Ed.).
Western North American Juniperus Communities: A Dynamic Vegetation Type. Springer-Verlag, New
York, NY, USA. 316 pp.
ESRI. 2018. ArcGIS Desktop: Release 10.6. Redlands, CA: Environmental Systems Research Institute.
Falkowski, M.J., J.S. Evens, D.E. Naugle, A.H. Khalyani, C.A. Hagen, A. Ponzanovic, and J.D. Maestas.
2017. Mapping tree canopy cover in support of proactive prairie grouse conservation in western North
America. Rangeland Ecology and Management 70:15–24.
Fick, W.H., and W.A. Geyer. 2010. Cut-stump treatment of Saltcedar (Tamarix ramosissima) on the Cimarron
National Grasslands. Transactions of the Kansas Academy of Science 113:223–226.
Frazier, I. 2018. The day the Great Plains burned. The New Yorker, 5 November 2018. Available online
at https://www.newyorker.com/magazine/2018/11/05/the-day-the-great-plains-burned. Accessed
1 July 2021.
Fryer, J.L., and P.S. Luensmann. 2012. Fire regimes of the conterminous United States. US Department of
Agriculture, Forest Service, Rocky Mountain Research Station, Fire Sciences Laboratory, Fire Effects
Information System (FEIS). Missoula, MT, USA. Available online at
www.fs.fed.us.database/feis/fire_regime_table/fire_regime_table.html. Accessed 1 July 2021.
Fuhlendorf, S.D., T.J. Hovick, R.D. Elmore, A.M. Tanner, D.M. Engle, and C.A. Davis. 2017. A hierarchical
perspective to woody plant encroachment for conservation of prairie-chickens. Rangeland Ecology
and Management 70:9–14.
Glasgow, S., and T.G. Bidwell. 2009. Management after wildfire. Oklahoma Cooperative Extension Service
Report NREM-2881. Stillwater, OK, USA. 4 pp.
Gleason, H.A. 1913. The relation of forest distribution and prairie fires in the middle west. Torrey Botanical
Society 13:173–181.
Glinski, R.L. 1977. Regeneration and distribution of sycamore and cottonwood trees along Sonoita Creek,
Santa Cruz County, Arizona. Pp. 116–123, In R.R. Johnson and D.A. Jones (Eds.). Importance, Preservation
and Management of Riparian Habitat: A Symposium. U.S. Department of Agriculture, Forest
Service, Rocky Mountain Forest and Range Experiment Station, General Technical Report RM-43.
Fort Collins, CO, USA. 217 pp.
Goldsmith, F.B., and N. Smart. 1982. Age, spacing and growth rate of Tamarix as an indication of lake
boundary fluctuations at Sebkhet Kelbia, Tunisia. Journal of Arid Environments 5:43–51.
Hatten, J.R. 2016. A satellite model of Southwestern Willow Flycatcher (Empidonax traillii extimus)
breeding habitat and a simulation of potential effects of tamarisk leaf beetles (Diorhabda spp.),
southwestern United States. U.S. Geological Survey Open-File Report 2016-1120. Reston, VA, USA.
88 pp.
Haukos, D.A., and C.W. Boal (Eds.). 2016. Ecology and Conservation of Lesser Prairie-Chickens. CRC
Press, Boca Raton, FL, USA. 294 pp.
Hohlt, J.C., B.J. Racher, J.B. Bryan, R.B. Mitchell, and C. Britton. 2002. Saltcedar response to prescribed
burning in New Mexico, In G.R. Wilde and L.M. Smith (Eds.). Research Highlights 2002: Range,
Wildlife, and Fisheries Management, Volume 33. Texas Tech University, College of Agricultural Sciences
and Natural Resources, Lubbock, TX, USA. 25 pp.
Joern, A., and K. Keeler (Eds.). 1995. The Changing Prairie: North American Grasslands. Oxford University
Press, New York, NY, USA. 244 pp.
Prairie Naturalist
M W. Sirch, D. S. Sullins, N. J. Parker, D. A. Haukos, J. D. Kraft, C. A. Hagen, K. A. Fricke
2022 54:11–23
22
Jones, S.R., and R.C. Cushman. 2004. A Field Guide to the North American Prairie. Houghton Mifflin
Company, New York, NY, USA. 510 pp.
Keeley, J.E. 2006. Fire management impacts on invasive plants in the western United States. Conservation
Biology 20:375–384.
Keeley, J.E. 2009. Fire intensity, fire severity and burn severity: A brief review and suggested usage. International
Journal of Wildland Fire 18:116–126.
Keeley, J.E., and P.W. Rundel. 2005. Fire and the Miocene expansion of C4 grasslands. Ecology Letters
8:683–690.
Korhonen, L., K.T. Korhonen, M. Rautiainen, and P. Stenberg. 2006. Estimation of forest canopy cover: A
comparison of field measurement techniques. Silva Fennica 40:577–588.
Lautenbach, J.D. 2017. The role of fire, microclimate, and vegetation in Lesser Prairie-chicken habitat
selection. M.Sc. Thesis, Kansas State University, Manhattan, KS, USA. 133 pp.
Lautenbach, J.M., R.T. Plumb, S.G. Robinson, C.A. Hagen, D.A. Haukos, and J.C. Pitman. 2017. Lesser
Prairie-chicken avoidance of trees in a grassland landscape. Rangeland Ecology and Management
70:78–86.
Lindley, T.T., D.A. Speheger, M.A. Day, G.P. Murdoch, B.R. Smith, N.J. Nauslar, and D.C. Daily. 2019.
Megafires on the southern Great Plains. Journal of Operational Meteorology 7:164–179.
McDaniel, K.C., and J.P. Taylor. 2003. Saltcedar recovery after herbicide-burn and mechanical clearing
practices. Journal of Range Management 56:439–445.
Miller, J.D., E.E. Knapp, C.H. Key, C.N. Skinner, C.J. Isbell, R.M. Creasy, and J.W. Sherlock. 2009. Calibration
and validation of the relative differenced Normalized Burn Ratio (RdNBR) to three measures
of fire severity in the Sierra Nevada and Klamath Mountains, California, USA. Remote Sensing of
Environment 113:645–656.
Monitoring Trends in Burn Severity. 2020. US Department of Agriculture, Forest Service, and US
Geological Survey. MTBS data access: National geospatial data. Available online at http://mtbs.gov/.
Accessed 3 May 2020.
National Oceanic and Atmospheric Administration (NOAA). 2020a. National Climatic Data Center, Climate
at a Glance. Available online at https://www.ncdc.noaa.gov/cag/county/time-series/KS-025/pcp/
ann/. Accessed 3 May 2020.
National Oceanic and Atmospheric Administration (NOAA). 2020b. National Climatic Data Center. Local
Climatological Data. Available online at https://www.ncdc.noaa.gov/cdo-web/datasets/LCD/stations/
WBAN:13985/detail. Accessed 3 May 2020.
Nippert, J.B., L. Telleria, P. Blackmore, J.H. Taylor, and R.C. O’Connor. 2021. Is a prescribed fire sufficient
to slow the spread of woody plants in an infrequently burned grassland? A case study in tallgrass
prairie. Rangeland Ecology and Management 78:79–89.
Noel, J.M., and N.L. Fowler. 2007. Effects of fire and neighboring trees on Ashe juniper. Rangeland Ecology
and Management 60:596–603.
Oberle, B., K. Ogle, A.E. Zanne, and C.W. Woodall. 2018. When a tree falls: Controls on wood decay
predict standing dead tree fall and new risks in changing forests. PLoS One 13:e0196712.
Owensby, C.E., K.R. Blan, B.J. Eaton, and O.G. Russ. 1973. Evaluation of Eastern Redcedar infestations
in the northern Kansas Flint Hills. Journal of Range Management 26:256–260.
Paull, D.A., J.W. Whitson, A.L. Marcotte, G.C. Liknes, D.M. Meneguzzo, and T.A. Kellerman. 2017.
High-resolution land cover of Kansas (2015). US Department of Agriculture, Forest Service, Research
Data Archive, updated 27 November 2017. Fort Collins, CO, USA. Available online at
https://www.fs.usda.gov/rds/archive/catalog/RDS-2017-0025. Accessed 3 May 2020.
Plumb, R.T., J.M. Lautenbach, S.G. Robinson, D.A. Haukos, V.L. Winder, C.A. Hagen, D.S. Sullins,
J.C. Pitman, and D.K. Dahlgren. 2019. Lesser Prairie-chicken space use in relation to anthropogenic
structures. Journal of Wildlife Management 83:216–230.
Polley, H.W., D.D. Briske, J.A. Morgan, K. Wolter, D.W. Bailey, and J.R. Brown. 2013. Climate change
and North American rangelands: Trends, projections, and implications. Rangeland Ecology and Management
66:493–511.
Pyne, S.J. 2017. The Great Plains: A Fire Survey. University of Arizona Press, Tucson, AZ, USA. 216 pp.
Prairie Naturalist
M W. Sirch, D. S. Sullins, N. J. Parker, D. A. Haukos, J. D. Kraft, C. A. Hagen, K. A. Fricke
2022 54:11–23
23
R Development Core Team. 2019. Program R version 3.6.1: A language and environment for statistical
computing. R Foundation for Statistical Computing. Vienna, Austria. Available online at https://
www.r-project.org/. Accessed 3 May 2020.
Racher, B.J., C.M. Britton, R.B. Mitchell, and D.B. Wester. 2003. Fire application to Saltcedar-dominated
riparian areas: Prescription development, hazardous fuels reduction, and ecosystem response. Annual
progress report to Joint Fire Science Program, Project 00-2-29. Phoenix, AZ, USA. 1 pp.
Ratajczak, Z., J.B. Nippert, J.M. Briggs, and S.L. Collins. 2012. Woody encroachment decreases diversity
across North American grasslands and savannas. Journal of Ecology 93:697–703.
Ratajczak, Z., J.B. Nippert, J.M. Briggs, and J.M. Blair. 2014. Fire dynamics distinguish grasslands,
shrublands and woodlands as alternative attractors in the central Great Plains of North America. Journal
of Ecology 102:1374–1385.
Rideout-Hanzak, S., D.B. Wester, C.M. Britton, and H.A. Whitlaw. 2011. Biomass not linked to perennial
grass mortality following severe wildfire in the Southern High Plains. Rangeland Ecology and
Management 64:47–55.
Rosenberg, K.V., A.M. Dokter, P.J. Blancher, J.R. Sauer, A.C. Smith, P.A. Smith, J.C. Stanton, A. Panjabi,
L. Helft, M. Parr, and P.P. Marra. 2019. Decline of the North American avifauna. Science 366:120–
124.
Samson, F.B., F.L. Knopf, and W.R. Ostlie. 2004. Great Plains ecosystems: Past, present, and future.
Wildlife Society Bulletin 32:6–15.
Stevens, L.E. 2002. Exotic tamarisk on the Colorado Plateau. Available online at https://web.archive.org/
web/20060206062038/http://www.cpluhna.nau.edu/Biota/tamarisk.htm. Accessed 1 July 2021.