RESEARCH ARTICLE ABSTRACT: Many forests, including the mixed-conifer forest of the Sierra Nevada, California, historically experienced a fire regime that generated considerable within- and among-fire environmental variability. Fire suppression has resulted in a heavier, more continuous fuel bed, which can cause today's prescribed fires to be considerably more homogeneous. To evaluate the potential importance of variability in fire severity on post-fire plant communities, I conducted an experiment to test whether understory species respond differently to sites burned under a heavy fuel load versus sites that burned under a light fuel • load. Woody fuel was added or removed from small forest plots in order to manipulate the fire severity during prescribed fire. The fuel load manipulations affected which species survived fire as well as which Fine-scale Patchiness species germinated after fire. Seven species (Chimaphila menziesii, Chrysolepis sempervirens, Osmorhiza chilensis, Pyrola picta, Phacelia hydrophylloides, Rubus parvijiorus, and Smilacina racemosa) were unable to survive fire in either treatment. Four species (Bromus laevipes, Galium sparsijiorum, Rubus in Fuel Load Can glaucifolius, and Symphoricarpos mollis) survived more often on sites that were burned under a light fuel load. Several fire-stimulated species (Calystegia malacophylla, Cryptantha sp., Gayophytum eriospermum, Solanum xanti, Arctostaphylos patula, Ceanothus parvifolius, and Lotus crassifolius) germinated after Influence Initial fire regardless of fuel load, but others (Claytonia perfoliata, Ceanothus cordulatus, Prunus emarginata, and Ribes tularense) appeared more often on the light fuel treatments. Seedlings of Abies concolor Post-fire Understory germinated more often on the sites that burned under heavy fuel conditions. The varying responses of different species suggest that small-scale variations in fuel load may cause heterogeneous patterns of surface fire severity, which in tum may contribute to maintaining floral diversity in the mixed-conifer Composition in forest understory. In order to conserve native understory plant diversity, fire management efforts to restore these forests might consider spatial heterogeneity in fire severity as a management objective. a Mixed Conifer Index terms: mixed-conifer, prescribed fire, Sierra Nevada, spatial heterogeneity, understory diversity Forest, Sequoia INTRODUCTION fire (Western Governors' Association 2001 ; National Park, Healthy Forests Restoration Act 2003). In many areas of the western United Our understanding of the effects of these California States, the suppression of natural fires has treatments on biodiversity and ecosystem produced heavy buildup of dead wood and functioning in these forests, while grow dense trees and has dramatically altered ing, remains incomplete. Recent studies on Monique E. Rocca 1 forest structure and composition. Fuel restoration methods have focused on the accumulations in affected forests are as importance of fire for restoring historical Department of Forest, Rangeland, sociated with both a rise in fuel biomass forest structure and for tree regeneration and Watershed Stewardship and an increased spatial continuity of fuels (Mutch and Parsons 1998; Mast et al. 1999; Colorado State University (Miller and Urban 2000a). Therefore, fires Stephenson 1999; Bailey and Covington Fort Collins, CO 80523-1472 in areas affected by fire suppression are 2002; Miller and Urban 2000b; Waltz et not only more intense, but they are also al. 2003; North et al. 2007). Several papers • likely to exhibit less within-fire patchiness have begun to shed light on herb and shrub than pre-European settlement fires would responses to restoration treatments (Collins have. For example, the extensive barren et al. 2007; Knapp et al. 2007; Korb et al. landscapes produced by intense wildfires, 2007; Wayman and North 2007). However, such as large areas of the Rodeo-Chediski isolating the effects of fire severity per se fire (Arizona, 2002) and the Hayman fire on understory plant communities has been (Colorado, 2002), are homogeneous at a difficult because fire severity is correlated large scale, and may not produce the range with factors such as canopy cover and soil of post-fire habitat conditions necessary moisture that affect both fire severity and for the reestablishment of all species. Less understory species distributions (North et attention has been given to the possibility al. 2005; Knapp et al. 2007). that prescribed fires, systematically ignited under mild weather conditions, might Here I define fire severity as the magnitude 1 Corresponding author: also lack the variability in fire severity of the ecological effects of fire. Under the [email protected]; 970-491-2112 required to restore the native understory condition that fuels are completely con community. sumed, higher fuel loads tend to result in higher fire severities relative to lighter fuel Efforts are currently underway to amelio loads because of the increased temperature Natural Areas Journal 29:126-132 rate heavy fuel conditions in western forests and duration of heat pulse into the soil, using mechanical thinning and prescribed greater fuel consumption, and increased
126 Natural Areas Journal Volume 29 (2), 2009 vegetation mortality (Reinhardt et al. 2001; of forest in the Tar Gap area of the East the effects of fire severity on post-fire Knapp and Keeley 2006). Variation in fire Fork of the Kaweah watershed in Sequoia community composition. Each 4-m x 4-m severity can create spatial heterogeneity in . National Park. Slopes were predominately site (block) was divided into four 2-m x the environment by generating gradients west facing at elevations between 2000 2-m quadrants (Figure la). Two quadrants in soil conditions, light availability, and and 2300 meters. The region experiences were randomly assigned to a "heavy" microclimate. This heterogeneity may a Mediterranean climate with cold, wet fuel treatment and two to a "light" fuel facilitate the establishment of a variety winters and warm, dry summers. Before treatment. For the quadrants in the heavy of specifically adapted species (Grubb European settlement, mixed-conifer for fuel treatment,woody fuel was added to 1977). Fire sensitive plants, for example, ests on west-facing slopes experienced a approximate the highest fuel loads ob may survive only on sites that were un mean fire return interval of about 14 years served in these forests, or approximately burned or lightly burned, whereas some (Kilgore and Taylor 1979), with most fires 10 kg dry woody fuel/m2, with a fuel size fire stimulated species might require high occurring in the late summer to fall dry class distribution roughly equivalent to fire temperatures to scarify and germinate. period (Caprio and Swetnam 1995). The that created by a fallen tree crown. Light Under fires of high temperature or dura study area had not burned since the Park fuel treatments were created by removing tion, seed banks may be killed such that started keeping fire records in 1921. all woody fuel in the lO-hour size class only plants able to disperse to the site after or larger (> 0.65 cm diameter), leaving fire, and tolerate its seedbed conditions, Abies concolor (Gord. & Glend.) Lindl. approximately 0.1 kg fine woody fuel/m2 are able to establish. If plants respond ex Hildebr. (white fir) is the dominant tree and enough litter to ensure the area would differently to variation in fire severity, species in this forest, with Pinus lamber burn mildly (Figure lb). a more heterogeneous burn should lead tiana Dougl. (sugar pine), and Calocedrus to a more diverse suite of species after decurrens [Torr.] Florin. (incense cedar) Understory plant composition was sur fire. If species diversity conservation is a occurring as less frequent components of veyed before the fire, in late July 2002. management goal, as it is in many areas the overs tory. The understory is composed Prior to fuel manipulations, all understory currently slated for restoration treatments, of a sparse yet diverse assemblage of shrub plant species were recorded within the learning how herbaceous diversity depends and herb species. In the absence of fire, central square meter of each 2-m x 2-m on fire-generated heterogeneity should be populations of many understory plants are treated quadrant (Figure 1a) leaving an a top priority. constrained to small areas within canopy unsampled buffer around the edges of each gaps, while other species have greatly treatment area. Newly germinated shrub Here I test the prediction that individual reduced adult populations compared to seedlings were counted, and other under species respond differently to fire severity the abundance of seed in the seed bank story species were recorded as present or in Sierra Nevada mixed-conifer forests. I (Rocca 2004). Many of the shrubs, includ absent. Seedlings of trees were tallied by conducted experimental manipulations of ing Lotus crassifolius [Benth.] Greene (big species. fuel loads on small forest floor areas to deervetch), Prunus emarginata [Dougl. ex produce burns that would release different Hook.] D. Diert. (bitter cherry), Ribes tular Sequoia National Park personnel ignited amounts of heat and create patches of dif the burn unit beginning on 10 October ferent fire severities. I added woody fuel ense [Coville] Fedde (sequoia gooseberry), to create high severity burning conditions Arctostaphylos patula Greene (greenleaf 2002. A strip headfire ignition pattern was and removed woody fuel to create light manzanita), and several Ceanothus species employed, resulting in a relatively complete severity burning conditions. I assessed (ceanothus) germinate from the seed bank burn. The year following the fire, in early plant community composition in each plot following fire, and are collectively referred August 2003, the vegetation surveys were before and one year after a prescribed fire. to as "fire-stimulated shrubs" hereafter. repeated. The 2003 growing season fol If species composition varies based on lowed a winter with average precipitation and a heavier than normal spring snowpack . pre-fire fuel loading, the argument could Experimental design and data due to late season storms and cool spring be made that spatial heterogeneity in fire collection severity is important for maintaining un temperatures (California Department of Water Resources; Rockwell et al. 2005). derstory diversity in this forest. In such a Within the study area, ten 4-m x 4-m case, homogeneous prescribed fires might sites were chosen for experimental fuel not adequately restore plant community manipulation. Each site was chosen to be Analyses composition. free of obstacles such as tree boles and rock outcrops (which would affect the area The objective of this study was to describe METHODS available for plant growth). All sites were the responses of individual species to fire covered by enough needle litter and fine severity; therefore, each species was ana woody fuel to carry fire, and were spaced lyzed separately using randomized-block Study area at least 50 meters apart. ANOVA with site as a blocking factor and fuel treatment as the primary factor (two The experimental fuel manipulations were A randomized-block experiment tested· replicates of each treatment per block). conducted within a 500-m x 1000-m area
Volume 29 (2), 2009 Natural Areas Journal 127 fire suggested that the fuel manipulations influenced fire severity, as measured by a) fuel consumption, in a predictable man 4 x 4 meter Site ner. All woody fuels in the treated blocks were thoroughly consumed, while some litter and fine fuels remained charred but ...... recognizable after fire in some of the light · . · . fuel areas. · 2 x 2 meter ...... quadrant · . · . RESULTS
Tree seedlings were rare in the sampled •• II II II II II II II liD II I!! !I I!! III I!I a area. A total of 61 seedlings were counted 1 m 2 understory · . · across all sites before the fire, and all of · . · subplot ·...... ·...... · } these were killed regardless of fire sever ity treatment. Because all seedlings died, only first-year germinants could occur post-fire. Sixteen first-year seedlings were counted after the fire. All were A. concolor b) seedlings except for one P. lambertiana seedling, which was not considered in the analysis. Thirteen A. concolor seedlings occurred on the heavy fuel quadrants while two occurred on the light fuel quadrants, suggesting that the more severely burned sites are the more suitable for A. con color germination (P = 0.0027, Poisson ANOVA).
Seedlings of six fire-stimulated shrub species had germinated one year after the fire (Table 1). Three species showed significant treatment preferences; Ceano thus cordulatus Kellogg (whitethorn), P. emarginata, and R. tularense preferred the light fuel sites (P =0.0018, P= 0.0055, and P < 0.0001 respectively, PoissonANOVA). Only one fire-stimulated shrub, R. tular fuel added fuel removed ense, occurred anywhere in the study area before fire; seedlings of this species were more likely to be found where the species Figure 1. Experimental design. a) Schematic of randomized block design. 4 x 4 meter sites were divided into 4 quadrants. Two quadrants were randomly assigned to a "heavy" fuel treatment and two to a occurred before the fire (P < 0.0001, Pois "light" fuel treatment. b) Photo of fuel manipulations. Both heavy fuel load treatments are on the left sonANOVA). side of the plot, with light fuel load treatments on the right. Twenty-one other (herb and non-fire For the species for which seedling counts determined using chi-squared tests. All stimulated shrub) species were observed were available (shrub and tree seedlings), analyses were conducted in S-plus 7.0 before and/or after the fire. Table 2 ar a general linear model with a log link (Insightful, Inc.). ranges all the herb, shrub, and tree species (i.e., Poisson regression) was fit. For the along a gradient from fire sensitive to fire herbaceous species, logistic regressions Each experimental site burned across its tolerant, along with results of statistical on presence/absence were run. A term for entire area, as evidenced by coverage of tests. Six species, Chimaphila menziesii pre-fire presence/absence was included ash and/or charred litter. While thermo (R. Br. ex D. Don) Spreng (little prince's in the models for each species that oc couple measurements of heat pulse into pine), Chrysolepis sempervirens (Kellogg) curred on the study sites before the fire. the soil were not available for this study, Hjelmqvist (bush chinquapin), Pyrola picta Significance of explanatory factors was visual inspection of the plots following Sm. (white-veined wintergreen), Phacelia
128 Natural Areas Journal Volume 29 (2),2009 ably burned for weeks to months. Over Table 1. Mean seeding count per quadrant, by treatment, for six fire-stimulated shrub species, with this time period, fires would at times creep p-values for a treatment effect. Species with significant treatment effects are highlighted in bold. or smolder and at other times run along the forest Ilooror flare up to the canopy. mean seedling count The resulting burn pattern would include Species Light fuel Heavy fuel p-value a variety of burn severities including un burned patches of ground, areas of steril greenleaf manzanita ized mineral soil, and occasional canopy (Arctostaphylos patula ) 1.05 0.55 0.075 gaps (Kilgore and Taylor 1979). If this snow bush kind of heterogeneous burn pattern was (Ceanothus cordulatus ) 0.35 0.00 0.00 typical during the evolutionary history of these species, the differential responses to littleleaf ceanothus fire severity observed in this study might (Ceanothus parvifolius ) 0.55 0.70 0.55 represent adaptations to different fire big deervetch generated microsites. Regardless of the (Lotus crassifolius ) 0.15 0.15 0.54 adaptive nature of differential responses to fire severity, this study suggests that bitter cherry a homogenous burn pattern, whether the (Prunus emarginata) 0.80 0.20 0.01 result of wildfire or a prescribed fire con sequoia gooseberry ducted under very dry moisture conditions (Ribes tularense) 2.25 0.70 <0.0001 (Knapp and Keeley 2.006), may lead to a loss of regeneration or survival micro sites for some species. hydrophylloides Torr. ex A. Gray (water and after fire. Apocynum androsaemifolium Based on the results of this experiment, one leaf phacelia), Rubus parviflorus Nutt. L. (bitter dogbane) appeared to survive could predict that homogeneously severe (thimbleberry), and Smilacina racemosa equally often after light fuel and heavy burns through heavy fuels are likely to favor (L.) Desf. (false Soloman's-seal), were fuel treatments, but new appearances oc shrubs and fire dependent annuals. Light found before the fire but disappeared curred more often on light fuel quadrants. burns may appear to support the highest completely after the fire, and one nearly Of the five fire-stimulated herbs, Claytonia post-fire species diversity because few spe disappeared (Osmorhiza chilensis Hook. peifoliata Donn. ex Willd. spp. perfoilata cies seem to explicitly require a severe sur &Arn. [mountain sweetcicely]). The other (miner's lettuce) appeared only on the light face fire environment. On the other hand, fourteen species were observed after the fire fuel sites after fire, while Calystegia mala burns that are uniformly low to moderate and were separated into three categories: cophylla (Greene) Munz ssp. malacophylla in severity might fail to generate the high (1) those that were present before the fire (Sierra morning-glory), Cryptantha sp. severity micro sites required by some tree ("fire survivors," eight species), (2) those (cryptantha species), Gayophytum erio seedlings. Studies suggest that Sequoia that represented appearances of new spe spermum Coville (Coville's groundsmoke), dendron giganteum [Lind!,] Buschh. (giant cies ("fire-stimulated herbs," five species), and Solanum xanti A. Gray (chaparral sequoia) germinates and survives b~st on and (3) those that were present before fire nightshade) germinated on both light and the ashy mineral soil created by the hottest but also often appeared in new places after heavy fuel load sites. No species, other fires (Harvey et al. 1980), and that seed fire (one species). than the tree seedlings of A. concolor, oc lings of many mixed-conifer tree species curred more often on the heavy fuel areas germinate mostoftenin the largest fire-cre Of the fire survivors, Bromus laevipes after the fire. ated canopy gaps (Kilgore 1973; Demetry Shear (woodland bromegrass), Galium and Duriscoe 1996). The observation that sparsiflorum W. Wight (Sequoia bedstraw), DISCUSSION one tree species, A. concolor, germinated Rubus glaucifolius Kellogg (glaucous more often on high severity burn sites leaved raspberry), and Symphoricarpos Species responded differently to fuel load, supports these previous studies; however, mollis Nutt. (creepingsnowberry) survived most likely in response to the different longer term observations are needed to more often on light fuel sites, while Arabis fire severities created by each treatment. determine whether germination from this repand~ S. Watson val'. repanda (Yosemite Species in the understory of this Sierra abundant and prolific seeder translates into rockcress), Pteridium aquilinUin (L.) Kuhn Nevada mixed conifer forest appear to take successful seedling establishment. I was (hairy brackenfern), Sambucus mexicana advantage of the variety of "regeneration unable to determine the best micro sites C. Presl ex DC. (common elderberry), and niches" (Grubb 1977) created by a patchy for tree regeneration for other tree spe Silene lemmonii S. Watson (Lemmon's sUlface fuel load (Table 2). Before Euro cies due to the small area sampled and, campion) were not fire sensitive and were pean settlement of California, fires ignited perhaps, because the prescribed burn did generally found in the same places before by lightning or by Native Americans prob- not generate the kind of intense fire that
Volume 29 (2), 2009 Natural Areas Journal 129 ..... w o Table 2. Understory species sorted along a gradient of fire sensitivity. z el- I:... e!. >... (!) FIRE SENSITIVE FIRE TOLERANT FIRE STIMULATED ~ rIl ~ '-o "' New occurrenc I: ...::s Survived more often on Post-fire occurrences New occurrences more New occurrences on both more often 0 e!. Disappeared after fire light fuel plots match pre-fire often on light fuel plots light and heavy fuel plots heavy fuel plo
little prince's pine woodland bromegrass* t bitter dogbane *# Sierra morning-glory white fir* (Chimaphila menziesii) (Bromus laevipes) (Apocynum androsaemifolium) (Calystegia malacophylla) (Abies concolor)
bush chinquapin Sequoia bedstraw' Yosemite rockcress t miner's lettucet cryptantha species (Chrysolepis sempervirens) (Galium sparsiflorum) (Arabis repanda) (Claytonia perfoliata) (Cryptantha sp.) t '0'" mountain sweetcicely§ glaucous-leaved raspbe~ hairy brackenfemt Coville's groundsmoke CIJ ~ (Osmorhiza chilensis) (Rubus glaucifolius) (Pteridium aquilinum) eriospermum) :.. ~.... white-veined wintergreen creeping snowberry* common elderbe~ chaparral nightshade o (Pyrola picta) (Symphoricarpos mollis) (Sambucus mexicana) (Solanum xanti) waterleaf phacelia Lenunon's campiont (Phacelia hydrophylloides) (Silene lemmonii) thimbleberry (Rubus parviflorus) false Soloman's-seal (Smilacina racemosa) '§'" snow bush * greenleaf manzamta. .E (Ceanothus cordulatus) (Arctostaphylos patula) 00 ] bitter cherry* littleleaf ceanothus -3 (Prunus emarginata) (Ceanothus parvifolius) 8 ~ Sequoia gooseberry* t bid deervetch t (Ribes tularense) (Lotus crassifolius) rz a- • significant treatment effect (p < 0.05) c 3 t significant pre-fire presence effect (p < 0.05) (!) t too few post-fire occurrences to analyze statistically <.0'" § survived on 5 out of 18 quadrants, no significant treatment effect
~ # displays characteristics of both columns o '"o <.0 creates canopy openings. species is also likely to persist for many Monique Rocca received her Ph.D. from years following fire, while the post-fire the Nicholas School of the Environment The species that establish after fire fall into spatial pattern of annuals and other read at Duke University. She is currently As ily dispersed species will be less apparent three general life history categories: (1) sistant Professor of Wildland Fire Science fire-stimulated shrubs (see Table 1), (2) an with increasing time-since-fire. at Colorado State University. nual forbs (c. perjoliata, G. eriospermum, and Cryptantha sp., and (3) perennial forbs A central justification for structural and (c. malacophylla, and S. xanti). Because process restoration goals is the assumption long-distance animal dispersal into the area that ecosystem health and biodiversity will is probably a relatively rare event given the follow from restoration of forest struc LITERATURE CITED burned forest understory (with sparse food ture or the reintroduction of ecosystem availability) and the short time since fire, processes such as fire (Landres et al. Bailey, J.D., and w.w. Covington. 2002. the animal dispersed shrub species likely 1999; Stephenson 1999). Under current Evaluating ponderosa pine regeneration rates following ecological restoration treatments germinated from the seed bank. At least management strategies, however, spatial in northern Arizona, USA. Forest Ecology two of the species showing reduced toler heterogeneity in fire effects is not stated as and Management 155:271-278. ances to more severe burn conditions, R. a process goal; nor is biodiversity preserva California Department of Water Resources. tularense and A. androsaemifolium, have tion stated as a structural goal (National California Data Exchange Center. Available smaller seeds than do many of the more Park Service 2004). If the results from this online
Volume 29 (2),2009 Natural Areas Journal 131 Keeley, J.E., B.A. Morton, A. Pedrosa, and Miller, C. and D.L. Urban. 1999. Forest hetero Rockwell, G.L., G.L. Pope, lR. Smithson, P. Trotter. 1985. Role of allelopathy, heat geneity and surface fire regimes. Canadian and L.A. Freeman. 2005. Water Resources and charred wood in the germination of Journal of Forest Research 29:202-212. Data-California, Water Year 2003, vol. 3. chaparral herbs and suffrutescents. Journal Miller, C., and D.L. Urban. 2000a. Connectiv Southern Central Valley Basins and The of Ecology 73:445-458. ity of forest fuels and surface fire regimes. Great Basin from Walker River to Truckee Kilgore, B.M. 1973. The ecological role of fire Landscape Ecology 15:145-154. River. U.S. Geological Survey, Water Re sources Division, California District, Sacra in Sierran conifer forests: its application to Miller, C., and D.L. Urban. 2000b. Modeling park management. Quaternary Research mento, Calif. Available online
132 Natural Areas Journal Volume 29 (2), 2009