Fire Science and Management

Contributed Paper
ROOM: Room 220 – Ruidoso

10:30AM Demographic Response of Northern Bobwhites to Pyric Herbivory of Non-Native Grassland
Eric D. Grahmann; Timothy E. Fulbright; Fidel Hernandez; Michael W. Hehman; David B. Wester; Alfonso Ortega-Santos; Blake A. Martin
Usable space for northern bobwhites (Colinus virginianus) has declined significantly over the last 3 decades in Texas because non-native grasses such as buffelgrass (Pennisetum ciliare) and Old World bluestems (Bothriochloa spp. and Dichanthium spp.) have replaced native vegetation. We hypothesized that burning patches in pastures dominated by buffelgrass followed by livestock grazing would increase limiting bare ground, traversable area by bobwhites, and forb and subshrub canopy cover, thereby increasing usable space and bobwhite population density. Our study was conducted during 2009-2011 in Southern Texas. Our experimental design was comprised of 2 blocks with 2, ±240-ha pastures (1 patch-burn and graze: treatment, and 1 graze only: control) in each. Pretreatment data were collected during 2009. We estimated grass standing crop in grazing exclosures (June‒September) and bare ground, traversable area, forb and subshrub cover, and herbaceous species richness along 120 transects during October 2009‒2011. Bobwhites were monitored using radio telemetry 2‒3 days/week from March-November. Bobwhite densities were estimated using 3 different methods. We compared means of vegetation attributes and bobwhite demographics between treatments and years. Means of vegetation community metrics were similar between treatments. However, grass standing crop tended to be lower in treatment pastures (June and August 2010 and September 2011 – 110.5 ± 26.2 g/m2) compared to control pastures (June and August 2010 and September 2011 – 145.5 ± 58.6 g/m2). Plant species richness was also greater (21%) in treatment pastures (4.6 ± 0.4/0.1m2) compared to control pastures (3.8 ± 0.4/0.1m2) (P ≥ 0.057). Beta scale heterogeneity was increased in treatment pastures for grass standing crop, bare ground, and herbaceous species richness. There was an increase in bobwhite densities in treatment pastures although other demographic metrics remained similar. Patch burning and grazing is a viable treatment option for restoring usable space and bobwhite populations in monotypic non-native grasslands.
10:50AM American Black Bear Movements in Response to Wildfire in Eastern Tennessee
Jessica Giacomini; Coy Blair; Joseph D. Clark; Lisa Muller
The American black bear (Ursus americanus) has been studied extensively throughout its southeastern range in the United States. Studies have focused on various aspects of black bear ecology including population size and structure, home-range metrics, habitat selection, movement patterns, denning preferences, and occurrence of human-bear conflicts. However, there has been no evaluation of how wildfire may affect American black bears in this part of their range. Multiple wildfires developed and spread across much of Sevier County, Tennessee from late November to early December 2016. Approximately 72-km2 were burned in the wildfires, including large areas in the city of Gatlinburg and Great Smoky Mountains National Park. Eight bears within the vicinity of the wildfires were equipped with Global Positioning System (GPS) radio-collars as part of 2 separate studies prior to the fire event. Collared bears included 3 yearling males, 2 adult males, and 3 adult females. We used a Before and After Control Impact (BACI) design to analyze three response variables: kernel utilization distribution areas of activity, movement rates, and mean direction of travel. We examined treatment-by-time interaction effects with treatment referring to bears occupying the area of the fires or bears outside of the area of the fire (controls) and three-day time periods before, during, and after the fire. Analysis of variance (ANOVA) tests were used to examine the differences between response variables and the effects of treatment, time, origin, and sex. We found no significant differences between bears using the wildfire area and those that were outside of the fire area (P>0.08). Origin (P<0.001) and sex (P<0.001) effects significantly impacted the movement rates of bears. Future studies may examine longer-term effects of wildfire on the movements of these GPS-collared bears as more data is collected.
11:10AM Building Consensus for Forest Restoration Through Wildfire Risk Assessment
Steven Bassett
Dry forests in the western United States are departed from their historically resilient condition, and are prone to large, high-intensity wildfires. These large, uncharacteristically intense wildfires can cause extensive damage to infrastructure, homes, fish and wildlife habitat, and other valued resources and assets. With insufficient funding to treat all departed forests, a strategy is needed to prioritize and optimize investment in restoration. With guidance from stakeholders in the fireshed surrounding the source watersheds of the San Juan-Chama trans-basin diversion project, The Nature Conservancy overlaid data characterizing the expected probability and intensity of fires, with the location and susceptibility of valued resources and assets, to identify the locations most at risk of wildfire. This uniform measure of wildfire risk was used to prioritize and optimize investment in forest restoration and fuels reduction treatments. Based on this fire modeling and spatial analysis, we found that: the values and locations that stakeholders thought had high risk, did indeed have high risk; sources watersheds have value to local communities as well as downstream communities; and risk is not distributed evenly across the landscape. Despite conflicting priorities from divergent stakeholder groups, the risk assessments and restoration strategy enabled coordinated investment across land ownership boundaries. The results may not be surprising but the process provided consensus on treatment prioritization based on stakeholder values, which will allow The Nature Conservancy and land management agencies to scale up forest restoration treatments with social license.
11:30AM Prescribed Rangeland Fire’s Influence on Northern Bobwhite Quail Chick Food Resources
Britt Smith; Robin Verble-Pearson; Brad Dabbert
Prescribed rangeland fire is a common management tool used to reduce woody plant dominance and improve forage production in the Texas Rolling Plains. Changes in vegetation structure and composition can influence the arthropod food resources for important game and wildlife species such as Northern Bobwhite Quail chicks. Bobwhite quail chicks require diets containing at least 20-25% protein, and optimally 28% protein, which comes primarily from the consumption of arthropods. Our study was conducted on two private ranches and one public wildlife management area in north central Texas. We applied two dormant season prescribed fires and opportunistically used a wildfire in 2014. In 2015 we conducted six dormant season prescribed fires. Ground-dwelling arthropods were sampled using five pitfall traps per each burned patch and each paired unburned site. Each 500 mL pitfall trap was filled with 200 mL of 1:1 solution of water and propylene glycol with a drop of unscented dish detergent. To sample vegetation-dwelling arthropods, we conducted sweep netting along two, 100 m transects per burned patch and paired unburned site. Arthropods collected in sweep nets were euthanized with ethyl acetate and frozen until they could be identified in the lab. Pitfall traps and sweep nets were collected weekly for four weeks between early July and early August in both years. We also measured foliar ground cover and plant species composition within 1 m2 square frames, and vegetation structure along four transects per paired burned and unburned sites. Arthropods were identified to highest taxonomic resolution possible. Arthropod taxa included in the analysis were truncated based on previous studies examining Northern Bobwhite Quail chick arthropod selectivity. Analysis will be conducted using model-based multivariate methods. Model covariates will include treatment, % bare ground, % live vegetation, Shannon’s H’ plant diversity, and vegetation visual obstruction.
11:50AM Wildfire Mapping and Damage Analysis in Grand Bay National Estuarine Research Reserve, Mississippi Using a Small Unmanned Aerial System with a Multispectral Payload
Sathishkumar Samiappan; Cary D. McCraine; Gray Turnage; Lee Hathcock; Robert Moorhead
Wildfires create a significant impact in terms of property damage and human safety, as well as ecosystem function by altering biodiversity. Resource managers require safe, easy to use, and cost-effective methods for quantifying damage caused by wildfires. Since wildfire events can create long-term ecosystem impacts, information on damage extents and impacted vegetation must be estimated in a timely manner for restoration and prediction purposes. Historically, mapping and damage estimation has been done by using a GPS unit and walking the boundary of a burned area, or using imagery from manned aerial or space-based platforms equipped with either a Medium Resolution Imaging Spectrometer or Advanced Very High Resolution Radiometer . These methods are time-consuming, expensive, and prone to human error. In this research, we demonstrate an alternative approach using small Unmanned Aerial Systems (UAS) equipped with high-resolution multispectral sensors to map and estimate wildfire damage at the Grand Bay National Estuarine Research Reserve (GBNERR) in Mississippi near the Gulf of Mexico. We collected approximately 4.5 sq. mi. of 5-band multispectral imagery (red, green, blue, red edge and near-infrared) using a PrecisionHawk Lancaster fixed-wing UAS platform in the GBNERR immediately after a wildfire in Feb. 2016. This approach negates cost issues and hazards of on-the-ground fieldwork . Imagery was collected at ~180 m. above ground level, with a ground sample distance of approximately ~ 6.3 cm., and used to create an orthomosaic. Airborne LiDAR data from Oct. 2016 and 4-band NAIP imagery from Dec . 2014 and Oct. 2016 were used to estimate pre- and post-burn damage and track the recovery process. Regression coefficients for comparison between radiance values were reported for blue, green, red and near-infrared bands. Damage and fire severity maps were produced using Normalized Difference Vegetation Index. Results show the efficiency of employing small UAS for mapping wildfires.


Contributed Paper
Location: Albuquerque Convention Center Date: September 26, 2017 Time: 10:30 am - 12:10 pm