Abstract
There are currently no camera trap studies that have taken place before and after a wildfire event. This presents an opportunity to quantify richness and to learn more about how mammal populations adjust to a post-fire environment. Using camera traps in a wash in the Red Cliffs Desert Reserve in southwest Utah, we collected data from January 7 to March 7 2020, and from August 30 to October 30 2020, for a total of 1830 camera trap nights both pre-and post-fire. Two data sets of mammalian populations were monitored, identified, quantified, and compared. This study found a 42% average decline in mammal sightings post-fire. Almost all species decreased in sighting numbers except in Merriam’s kangaroo rats (Dipodomys merriami) and kit foxes (Vulpes macrotis), which increased in numbers by 18% and 80%, respectively. This study found that 3 months after a wildfire event, mammal sightings declined 42% from pre-wildfire numbers.
INTRODUCTION
Wildfires are an integral part of ecosystem succession, but have substantially increased recently due to human-caused climate change and higher population densities (Brehme et al., 2011). Wildfires can have long- and short-term effects on mammal populations, and these effects increase with the size and frequency of wildfire in an area (Banks et al., 2010). On July 13, 2020, a 11,993-acre wildfire burned the Turkey Farm area of the Red Cliffs Desert Reserve in St. George, Utah (Kellam, 2020).
Our team set up camera traps in the area in January and in August of 2020 for two-month periods to collect data on wild mammal populations. The aforementioned Turkey Farm wildfire burned the area in July of that year. To date, no camera trap studies exist which have collected data on mammal species richness pre- and post-wildfire. The purpose of this study was to determine whether wildfire has an impact on mammal populations in the desert southwest.
METHODS
The area of study spans approximately 275 acres in a steep igneous rock and sandstone valley in the Turkey Farm area of the Red Cliffs Desert Reserve near St. George, Utah (Figs. 1a, b). It is a sandy-bottom wash, edged with shrubs and bushes. Common species of plants in the area of study are desert almond (Prunus fasciculata), creosote (Larrea tridentata), desert sage (Artemisia filifolia), and four-wing saltbush (Atriplex canescens).
It is home to carnivores, such as coyotes (Canis latrans), and bobcats (Lynx rufus). It is also home to herbivores and omnivores such as kangaroo rats (Dipodomys merriami), white-tailed antelope squirrel (Ammospermophilus leucurus), and black-tailed jackrabbits (Lepus californicus). The valley provides occasional water from snowmelt and rainstorms and serves as a geographic funnel for wild mammals. These geographic features suggested that a higher potential density of wildlife sightings were likely in the area. The canyon is in a remote part of the reserve and, as such, sees very little human activity. Initially, these were the reasons that Turkey Farm wash was chosen as the area of study for camera-trap data collection, but then the Turkey Farm wildfire changed the ecosystem.
Species occurrence data were recorded pre- and post-fire via 15 Browning Strike Force trail cameras from January 7 to March 7, 2020 and from August 30 to October 30, 2020. Motion sensor and infrared technology allowed data to be collected day and night for both two-month-long periods. Cameras were set up approximately 1m. off the ground on aluminum stakes, 200m apart from one another on the banks of the wash. Cameras were checked every 2wk. for proper functioning. At the end of the data collection period, photo events of wildlife activity were entered into the trail camera software Wild.id.
RESULTS
This study recorded an approximate 42% decrease in mammalian sightings between pre- and post-fire numbers. In total, 16 species of mammal were recorded. More than 2100 photo events were documented and sorted (Fig. 4). The species accumulation curves demonstrated a much steeper slope in the pre-fire data collection period than in the post-fire period (Fig. 3). The first 12 species were observed in the first 12 days in the pre-fire period and in 21 days in the post-fire period. The post-fire period data contained a lower number of wild mammal sightings except for merriam’s kangaroo rats and kit foxes (Fig. 4).
Five species were chosen as focal species due to their high density in the area as well as their diversity of niches in the ecosystem: white-tailed antelope squirrel, black-tailed jackrabbit, bobcat, coyote, and Merriam's kangaroo rat.
Some, such as black-tailed jackrabbits, shifted activity patterns very little, whereas the carnivores of the focal species, bobcats and coyotes, showed a more apparent preference for dusk in patterns of activity (Fig. 2).
Black-tailed jackrabbits sightings decreased 47% in the area after the wildfire. White-tailed antelope squirrels decreased by 40% in sighting numbers. Bobcat sightings decreased 36% post-wildfire, and bobcat activity changed; ranging between 17:00 to 7:00 pre-fire to being active at 21:00 and 1:00 almost exclusively, post-fire (Fig. 2). Deer mouse sightings decreased by 92% (Fig. 4).
Merriam’s kangaroo rat sightings increased 18% post-wildfire, and activity patterns shifted minimally (Figs. 2, 4). Kit fox sightings increased 80% in the post- wildfire period (Fig. 4).
DISCUSSION
Individual Species Responses
Black-tailed jackrabbits are herbivores which rely on perennial grasses for many of their calories (Johnson and Anderson, 1984). Many perennial grasses did not survive the fire and were slow to grow back. As a result, the ecosystem no longer had the carrying capacity for grass-eating mammals like rabbits and sightings declined. Black-tailed jackrabbit activity patterns did not shift very notably. Jackrabbits were consistently sighted during the sunset hour and an hour or two before sunrise, pre-and post fire.
White-tailed antelope squirrels feed primarily on foliage, though they have been noted feeding on seeds, depending on food scarcity and time of year (Letnic et al., 2013). These squirrels were the most commonly sighted mammals and were closest to the total average for pre- and post-fire numbers at approximately 40%. White-tailed antelope squirrel activity patterns shifted very little, probably due to seasonal change. In pre-fire data, they were often sighted an hour or two hours before sunset or after sunrise, whereas post-fire they were often sighted later in the morning and earlier in the afternoon. White-tailed antelope squirrels sightings decreased dramatically in the 1:00pm hour across both data sets.
Bobcats are the biggest carnivore in this study and are so large that they typically have no choice but to flee in wildfire events due to lack of appropriate protection from fire (van Mantgem et al., 2015). Wildfire also renders an area difficult to hunt in for large predators, as it reduces prey numbers and areas of cover for stalking prey (Pausas, 2018). Before the wildfire, bobcats were seen many hours of the night. After the fire, sightings happened almost exclusively in the 9:00 hour, close to peak time for black-tailed jackrabbits. The homogenous nature of sightings after the wildfire may be due to a reduction of population in the area down to a single individual with well-set patterns for hunting, and thus, is seen often at the same time in the same place. Though this study did not make note of individuals.
Coyotes showed very similar patterns to the bobcat and likely suffered from the same issues in hunting and prey density due to filling similar niches in the ecosystem (Cunningham et al., 2006). Coyotes went from being sighted all hours of the night to being sighted almost exclusively at sunset, perhaps the time they set out to hunt for the night.
Kit fox numbers increased by 80%; they likely benefited from the decline of the coyotes and bobcats in a trophic cascade effect. The larger carnivores left the area in the event of wildfire, and the area was made less capable of supporting them. Kit foxes live in dens, however, so in the case of wildfire they can take shelter underground (Smith et al., 2013). They are also considerably smaller than bobcats and coyotes and so have access to smaller rodents such as white-tailed antelope squirrels, for dietary needs. Kit foxes likely did not have to relocate due to absence of larger carnivores in the area creating enough prey and hunting time.
Merriam’s kangaroo rats live in burrows as well, and also increased in sightings in the post-fire numbers (van Mantgem et al., 2015). Dens provide shelter from fire, as well as ample water. Merriam’s kangaroo rats eat mainly creosote seeds and insects. It has been observed that even when water is abundant they drink only rarely, instead getting their water from dietary foliage and seeds eaten after being left in burrows to absorb water from the moist air underground (Nagy and Gruchacs, 1994). A large unburned patch of creosote brush near the study site may have fueled local kangaroo rat diets. The success of kangaroo rats in a post-wildfire environment has been studied previously and they are known to do well (Brehme et al., 2011, Banks et al., 2010). Merriam’s kangaroo rat activity patterns changed minimally from pre- to post-fire numbers.
Two main systems that mammals can use to protect themselves from wildfire are refugia and avoidance (Pausas, 2018). Larger mammals typically have no option but to use the strategy of avoidance and flee the area. Many smaller mammals seek refuge in burrows, around rocks, or in riparian zones in the event of wildfire. The two species that did well post-wildfire are nocturnal and thus readily have spaces available for refuge from a wildfire. Nocturnal mammals exhibit higher survivorship of wildfire events than diurnal mammals due to their habits of residing in dark places, unexposed from fire (Smith, 2013). The increase in sightings of Merriam’s kangaroo rats and kit foxes supports this body of research.
Overall Response to Wildfire
The results of this study indicate that southern Utah mammal populations exhibited a strong response to wildfire. Decreases in total sightings were 42% in the post-wildfire data. Larger carnivores such as bobcats, and their prey; grass-dependent mammals such as black-tailed jackrabbits and white-tailed antelope squirrels, whose diet consists of 20-60% foliage, decreased similarly with average numbers (Belk, 1991).
In environments where wildfire is becoming more common, it’s possible that nocturnal, burrowing creatures will begin to dominate ecosystems (Smith et al., 2013) and large carnivores will be forced to expand their range to find food and avoid burned areas (van Mantgem et al., 2015). This study may be helpful in more effectively understanding mammalian interaction such as trophic cascading with wildfire in an era where wildfire is becoming more common.
REFERENCES CITED
Banks, S., Dujardin, M., McBurney, L., Barker, M., Lindenmayer, D., 2011. Starting points for small mammal population recovery after wildfire: recolonisation or residual populations? Oikos Vol. 120, No. 1: 26-37.
Belk, M. and Smith, D. 1991. Ammospermophilus leucurus. Mammalian Species N0. 368: 1-8.
Brehme, C., Clark, D., Rochester, C. et al. 2011. Wildfires alter rodent community structure across four vegetation types in Southern California, USA. Fire Ecology 7: 81–98.
Cunningham, S., Kirkendall, L., Ballard, W., 2006. Gray fox and coyote abundance and diet responses after a wildfire in central Arizona. Western North American Naturalist 66(2): 169-180.
Heilbrun, R., Silvy, N., Peterson, M., and Tewes, M. 2006. Estimating bobcat abundance using automatically triggered cameras. Wildlife Society Bulletin 34: 69-73.
Kellam, J. 2020. Red Cliffs National Conservation Area Cottonwood Trail Fire Tortoise Mortality Survey Report. St George Utah, Bureau of Land Management, 12 pp.
Letnic, M., Tischler, M. and Gordon, C. 2013. Effects of fire on desert small mammals. Austral Ecology 38: 841-849.
Nagy, K. and Gruchacz, M. 1994. Seasonal water and energy metabolism of the desert-dwelling kangaroo rat (Dipodomys merriami). Physiological Zoology 67(6): 1461–1478.
Schuette P., Diffendorfer J., Deutschman D., Tremor S., Spencer W. 2014. Carnivore distributions across chaparral habitats exposed to wildfire and rural housing in southern California. International Journal of Wildland Fire 23: 591-600.
Smith, A., Bull, C. and Driscoll, D. 2013. Successional specialization in a reptile community cautions against widespread planned burning and complete fire suppression. J Appl Ecol, 50: 1178-1186.
van Mantgem, E., Keeley, J. and Witter, M. 2015. Faunal Responses to Fire in Chaparral and Sage Scrub in California, USA. Fire Ecology 11: 128–148.
Figure Captions
Figure 1a A satellite map of the approximately 270 acre area of study in the Turkey Farm area of the Red Cliffs Desert Reserve near St George, UT.
Figure 1b A topographic map of the approximately 270 acre area of study in the Turkey Farm area of the Red Cliffs Desert Reserve near St George, UT.
Figure 2a 24 hour activity chart for bobcats from the Turkey Farm Wash area in the Red Cliffs Desert Reserve, near St. George UT, before and after the 2020 wildfire.
Figure 2b 24 hour activity chart for coyotes from the Turkey Farm Wash area in the Red Cliffs Desert Reserve, near St. George UT, before and after the 2020 wildfire.
Figure 2c 24 hour activity chart for white-tailed antelope squirrels from the Turkey Farm Wash area in the Red Cliffs Desert Reserve, near St. George UT, before and after the 2020 wildfire.
Figure 2d 24 hour activity chart for Merriam's kangaroo rats from the Turkey Farm Wash area in the Red Cliffs Desert Reserve, near St. George UT, before and after the 2020 wildfire.
Figure 2e 24 hour activity chart for black-tailed jackrabbits from the Turkey Farm Wash area in the Red Cliffs Desert Reserve, near St. George UT, before and after the 2020 wildfire.
Figure 3 Species accumulation curves of the camera trap survey at the Turkey Farm Wash area in the Red Cliffs Desert Reserve, near St. George UT, before and after the 2020 wildfire.
Figure 4 Overall species sighting counts from the Turkey Farm Wash area in the Red Cliffs Desert Reserve, near St. George UT, before and after the 2020 wildfire.
Figure 1a
Figure 1b
Figure 2a
Figure 2b
Figure 2c
Figure 2d
Figure 2e
Figure 3
Figure 4
