Citation:
Fryer, Janet L., compiler. 2021. Research Project Summary: Effects of wildfire on a mixed-deciduous riparian community in southeastern Arizona. In: Fire Effects Information System, [Online]. U.S. Department of Agriculture, Forest Service, Rocky Mountain Research Station, Missoula Fire Sciences Laboratory (Producer). Available: https://www.fs.usda.gov/database/feis/research_project_summaries/Bock2014/all.html [].

Fire damage to five riparian tree species was evaluated in a southeastern Arizona mixed-deciduous riparian community burned by wildfires in 2002, 2009, or both. Trees were measured in the summers of 2003–2004 and 2012–2013. On sites burned by one fire, percentage of individuals killed was 36% for Fremont cottonwood, 12% for velvet ash, 7% for Arizona walnut, 5% for desert-willow, and 0% for Arizona sycamore; while percentage of individuals top-killed and sprouting was 59% for desert-willow, 41% for Arizona walnut, 25% for Arizona sycamore, 21% for Fremont cottonwood, and 20% for velvet ash. Despite having relatively low fire severity, the second fire did substantial damage to trees on the site burned by both fires. By 2012–2013, 26% of the 250 trees had died, 32% survived only as relatively small sprouts from top-killed trees, and only 22% survived with little to no fire damage.

This study suggests that among mature trees of these five species, Arizona sycamores and velvet ashes are most likely to survive fire, desert-willows are most likely to be top-killed and survive as sprouts, and Fremont cottonwoods are most vulnerable to fire kill. The authors concluded that a single fire can substantially reduce both the abundance and crown volume of mature riparian trees in these communities, and a second fire soon after the first (in this case, 7 years after) is likely to increase rates of fire kill or top-kill of these species.

This document summarizes information from a research project conducted from 2003–2013 in a mixed-deciduous riparian community in southeastern Arizona. FEIS's Research Project Summaries are intended to provide concise information about the effects of a particular fire (or fires) on a specific plant community, and to supplement FEIS Species Reviews with detailed information on these fires and their effects in a particular location. The studies summarized are selected based on their integration of fire effects information, and they have relatively complete descriptions of the burning conditions, fire weather, and fire behavior; and on the effects of fire on vegetation (i.e., burned and unburned, pre- and postfire, or first and second fire data).

Common names are used throughout this summary. For a complete list of the common and scientific names of species mentioned in this summary and for links to FEIS Species Reviews, see the Appendix.

Sources:
Unless otherwise cited, the information in this Research Project Summary comes from the following paper:

Bock, Carl E.; Bock, Jane H. 2014. Effects of wildfire on riparian trees in southeastern Arizona. The Southwestern Naturalist. 59(4): 568-574 [2].

This study took place on the Appleton–Whittell Research Ranch, a 3,160-ha sanctuary of the National Audubon Society in the Sonoita Valley, Santa Cruz County, Arizona (31° 39' N, 110° 32' W). Five tree species (table 1) were sampled in three canyons of the San Pedro River Watershed that had burned in the 2002 Ryan Fire, the 2009 Canelo Fire, or both wildfires [2].

The study area consists of mesas, slopes, and low benches. These are adjacent to ephemeral drainages that drain into the three canyons. Elevation averages around 1,500 m. Temperatures range from a daily January minimum mean of 23 °C to a daily June maximum mean of 33 °C. Mean annual precipitation is 43 cm, about 60% of which occurs during monsoons in July and August [2].

Each of the three canyons supports mixed-deciduous riparian woodlands codominated by Arizona walnut, Fremont cottonwood, desert-willow, velvet ash, and Arizona sycamore, in descending order of frequency. Goodding’s willow is an associated tree. Warm-season perennial bunchgrasses dominate the surface vegetation (e.g., deergrass and saltgrass). Earlier fires might have killed trees that were particularly vulnerable to fire damage because of their condition or position in the floodplain, reducing stand density prior to the 2002 and 2009 fires. Vegetation in the upland areas consists of encinal woodlands codominated by Arizona oak, Emory oak, and velvet mesquite, with perennial bunchgrasses in the surface layer [2].

Data on prefire woody fuel loads and spatial variation of prefire fuels were unavailable. The study site had abundant fine fuels because it had not been grazed by livestock since 1968. Part of the study area (Upper O'Donnell Canyon) was burned by wildfires in both 1975 and 1987, but fine fuels "almost certainly had recovered by 2002" [2].

Determining historical fire intervals in mixed-deciduous riparian communities is difficult because the hardwood overstory species are susceptible to fungal heart rot, which eliminates the fire scar records [7]. Limited research (e.g., [3,4,5], review by [9]) and models (e.g., [6]) suggest that fire might have been historically infrequent in these communities.

Plant communities at these study sites are part of the Biophysical Setting (BpS) described in table 2.

Information on plant species' phenological stage at the time of the wildfires was not available in the published study. The tree species were likely leafed out and beginning to flower (see the Seasonal Development sections in the tree Species Reviews linked to in the Appendix).

Ryan wildfire: Spring, moderate to high severity
Canelo wildfire: Spring, low severity

Satellite images analyzed as part of the interagency Monitoring Trends in Burn Severity project (www.mtbs.gov) showed few if any unburned pixels within the perimeter of either fire. Imagery from the 2002 Ryan fire showed about equal mixes of moderate- and high-severity pixels in the canyons where trees were sampled. Imagery from the 2009 Canelo fire showed nearly all low-severity pixels [2].

The 1,625-ha Canelo Fire ignited on 5 May 2009, also south of the ranch. It burned north through Lyle and Upper O’Donnell canyons and their tributaries on 5–6 May 2009. Maximum temperature on 6 May was 29.0 °C, with sustained winds up to 17 km/h in late afternoon. The Canelo Fire was contained, in part, by applying a backfire before it spread to Post or Lower O’Donnell canyons [2].

The Ryan Fire was more severe than the Canelo fire. This may have been because fuels were drier in 2002 than in 2009 and because woody fuel loads were likely higher during the first fire than in areas that were reburned in the second fire (i.e., all areas except Upper Lyle Canyon). The Ryan Fire followed an unseasonal drought in which only 5 mm of precipitation fell in the prior 3 months. In contrast, 185 mm of precipitation fell in the 3 months prior to the Canelo Fire [2].

The wildfires substantially reduced abundance and crown volume of mature riparian trees in the study area. After one fire, 13% of the 250 trees were dead and 31% were top-killed with relatively small sprouts. The second fire burned 120 of the trees sampled that were burned by the first fire (table 1). After one or two fires, 26% of the trees were dead and 32% were top-killed with sprouts (table 3). Data were insufficient for comparing the effects of one fire to those of two fires. Significance level was set at P ≤ 0.05 for this study [2].

Types and amounts of fire damage differed among individuals and species. Fremont cottonwood and desert-willow had the lowest average condition ranks. For Fremont Cottonwood, mortality was greatest (36%) and postfire sprouting least, after one fire. In contrast, many desert-willows (59%) were top-killed and sprouted. Arizona walnut was intermediate in nearly every category of fire damage. A greater proportion of velvet ashes and Arizona sycamores survived with little to no damage compared to the other tree species, and they had the highest average postfire condition ranks after one fire (table 3, fig. 1) [2].

Figure 1—Mean (SE) postfire condition rank for five riparian trees after one fire in either 2002 or 2009 (open bars), and after one or both fires (closed bars). Reproduced from Bock and Bock (2014) [2], used with permission.

The 2009 fire did "substantial" damage to the trees in Upper O’Donnell Canyon despite its relatively low severity. The authors suggested that the affected trees were already weakened by the 2002 fire. On this twice-burned site, Fremont cottonwoods and desert-willows appeared to be in the worst condition, and Arizona sycamores in the best condition, in 2012–2013. The condition of velvet ashes was intermediate to these, but they were in worse condition than after one fire. Many velvet ashes that survived the 2002 fire in relatively good condition were killed or top-killed by the 2009 fire (table 3, fig. 1) [2].

Mean postfire basal trunk diameter differed among the riparian tree species (table 4). Trunk basal diameter was positively associated with condition rank after one fire for all trees combined and individually for Arizona walnut, Fremont cottonwood, and desert-willow. However, trunk basal diameter was not associated with condition rank for the two species least damaged by fire: Arizona sycamore and velvet ash [2].

Mean number of postfire sprouts per tree and mean sprout diameter differed among species (table 4). While Fremont cottonwood was least likely to sprout after fire (table 3), its sprouts were larger on average than sprouts of the other tree species [2].

Fire is a natural disturbance in riparian ecosystems, and it provides many important ecological benefits [9]. Although historical fire frequency is uncertain [2,8,9], mixed-deciduous riparian communities are high-priority areas for management, and maintaining mature trees is important for conservation of wildlife and fishery habitats, erosion control, protecting water quality, and providing aesthetic and recreational value [2,9].

This study suggests that for midelevation riparian areas of the Southwest, a single fire can substantially reduce both the abundance and crown volume of large deciduous trees, and a subsequent fire soon after the first fire (in this case, 7 years after) can result in high mortality rates of mature trees. Of the 250 trees originally tagged and located in 2003–2004, 58% had either died from fire or survived only as relatively small ground-level sprouts by 2012–2013. Among the five species studied, Arizona sycamore and velvet ash were most likely to survive fire as mature trees, desert-willow was most likely to survive as sprouts, and Fremont cottonwood was most vulnerable to fire kill [2].

REFERENCES:

1. Barrett, S.; Havlina, D.; Jones, J.; Hann, W.; Frame, C.; Hamilton, D.; Schon, K.; Demeo, T.; Hutter, L.; Menakis, J. 2010. Interagency fire regime condition class guidebook (FRCC), [Online], (Version 3.0). In: Interagency fire regime condition class website. U.S. Department of Agriculture, Forest Service; U.S. Department of the Interior; The Nature Conservancy (Producers). Available: https://www.landfire.gov/frcc/frcc_guidebooks.php [2021, February 2]. [85876]

2. Bock, Carl E.; Bock, Jane H. 2014. Effects of wildfire on riparian trees in southeastern Arizona. The Southwestern Naturalist. 59(4): 568-574. [89514]

3. Busch, David E.; Smith, Stanley D. 1993. Effects of fire on water and salinity relations of riparian woody taxa. Oecologia. 94(2): 186-194. [88320]

4. Dwire, Kathleen A.; Kauffman, J. Boone. 2003. Fire and riparian ecosystems in landscapes of the western USA. In: Young, Michael K.; Gresswell, Robert E.; Luce, Charles H., eds. Selected papers from an international symposium on effects of wildland fire on aquatic ecosystems in the western USA; 2002 April 22-24; Boise, ID. In: Forest Ecology and Management. 178(1-2): 61-74. [44923]

5. Lambert, Adam M.; D'Antonio, Carla M.; Dudley, Tom L. 2010. Invasive species and fire in California ecosystems. Fremontia. 38(2): 29-36. [89006]

6. LANDFIRE. 2016. Biophysical settings layers (LF 2016 Remap), [Online]. In: LANDFIRE. U.S. Department of the Interior, Geological Survey; U.S. Department of Agriculture (Producers). Available: https://landfire.gov/bps.php. [96126]

7. Swetnam, Thomas W.; Baisan, Christopher H. 1996. Fire histories of montane forests in the Madrean Borderlands. In: Ffolliott, Peter F.; DeBano, Leonard F.; Baker, Malchus B., Jr.; Gottfried, Gerald J.; Solis-Garza, Gilberto; Edminster, Carleton B.; Neary, Daniel G.; Allen, Larry S.; Hamre, R. H., tech. coords. Effects of fire on Madrean Province ecosystems: A symposium proceedings; 1996 March 11-15; Tucson, AZ. Gen. Tech. Rep. RM-GTR-289. Fort Collins, CO: U.S. Department of Agriculture, Forest Service, Rocky Mountain Forest and Range Experiment Station: 15-36. [28061]

8. Turner, Raymond M. 1974. Quantitative and historical evidence of vegetation changes along the upper Gila River, Arizona. Gila River Phreatophyte Project; USGS Professional Paper 655-H. Washington, DC: U.S. Govt. Print. Off. 19 p. [36381]

9. Webb, Amanda D.; Falk, Donald A.; Finch, Deborah M. 2019. Fire ecology and management in lowland riparian ecosystems of the southwestern United States and northern Mexico. Gen. Tech. Rep. RMRS-GTR-401. Fort Collins, CO: U.S. Department of Agriculture, Forest Service, Rocky Mountain Research Station. 132 p. [95328]