Postfire Recovery of Chaparral along an Elevational Gradient in Southern California

Abstract

Recovery of chaparral after wildfires was studied at five sites along an elevational gradient from 300 to 1550 m. In January, two months after fire, shrubs at all sites had begun resprouting but no shrub seedlings were present. Shrub resprouts were more abundant at the higher elevations although by June average shrub size was smaller at the higher than at the lower elevations. Initial shrub seedling densities were much higher than often reported for southern California chaparral and was probably due to sampling early in the growing season. Estimates of seedling mortality from March to June ranged from 68% to 98% and was dependent on the species. Herbaceous vegetation was most extensively developed at the lowest elevation site. At some sites, herbaceous perennials were a major part of the cover and these species typically established early in the season, largely from resprouts. Annuals were abundant at all but the highest elevation site and non-native annaul species dominated the flora at the lowest elevation site. Chaparral is a shrub vegetation distributed in various elevational belts throughout parts of the southwestern U.S. though it is most extensively developed in California. This west coast component is floristically distinct from chaparral in other regions (Daubenmire, 1943). Particularly noteworthy is the dominance of California chaparral by the shrub Adenostoma fasciculatum, chamise, (names according to Munz, 1974) and the absence of this species in the interior. Wildfires are common in California and chaparral vegetation has a number of characteristics which have been interpreted as evolutionary responses to frequent fires. Many of the shrub species, e.g., A. fasciculatum, some Arctostaphylos (manzanita) spp. and some Ceanothus spp., possess specialized lignotubers which rapidly regenerate aboveground parts after fire and a number of species produce seeds which require fire for germination. Associated with the shrubs is a diverse herbaceous flora which is sparce in the mature vegetation but proliferates after fire. Some of the species, termed 'pyrophyte endemics' (Hanes, 1977), may dominate the first year after fire and then disappear for decades until the next fire. Previous studies have provided a great deal of information on post-fire successional changes in both the shrub and herb floras (e.g., Horton and Kraebel, 1955; Hanes, 1971; Vogl and Schorr, 1972; Keeley et al., 1981; Zedler et al., 1983). Typically all aboveground plant biomass is killed by fire. In the first spring after fire annual herbs dominate. These herbs arise from previously dormant seeds in the soil and few herbs colonize burned sites from the outside. The first growing season is also the time of establishment for the vast majority of perennial species which will dominate in later years. Short-lived suffrutescent species germinate profusely but usually do not dominate sites until the third or fourth years after fire. T E SOUTHWESTERN NATURALIST 31(2):177-184 MAY 22, 1986 This content downloaded from 207.46.13.127 on Fri, 14 Oct 2016 04:29:42 UTC All use subject to http://about.jstor.org/terms The Southwestern Naturalist Certain species, e.g., A. fasciculatum, resprout from basal burls on burned plants or establish seedlings from soil-stored seed. Some species, e.g., Quercus dumosa (scrub oak) resprout but seldom establish seedlings, and other species, such as certain Arctostaphylos spp. and Ceanothus spp., are incapable of resprouting but establish abundant seedlings in the first growing season after fire. The purpose of this study was to provide further information about the following aspects of chaparral community post-fire recovery: (i) patterns of recovery along an elevational gradient, (ii) rates of recovery beginning in winter of the first post-fire year, (iii) differences in recovery rate and elevational distribution between sprouting and seeding life history types for both herbs and shrubs. STUDY SITES-Characteristics of the study sites are shown in Table 1. The two lower sites were in the Verdugo Hills northwest of Los Angeles (Los Angeles Co.) and the other three were on Mt. Baldy north of Pomona (site III was in Los Angeles Co., sites IV and V were in San Bernardino Co.). All sites burned in November 1980 and as is typical of California chaparral fires, all aboveground biomass was killed. Based on the size of shrub remains, it was estimated that all sites had been free of fire for at least 15 years prior to burning. Local and federal agencies helicopter seeded much of this burned area soon after the fires with annual ryegrass (Lolium perenne spp. multiflorum). All sites have a mediterranean climate with precipitation largely absent between April and October. Annual rainfall averaged between 375 and 750 mm from the lowest to the highest elevation site. Summer temperatures are similar across this gradient; July average temperatures are 23°C and 22°C for the lowest and highest sites, respectively. However, winter temperatures are markedly different; 12°C vs 4°C for the average January temperature at the lowest and highest sites (NOAA, 1981). Soils at all sites were shallow, coarse to moderately textured, and rocky. METHODS-All sites were first sampled in January 1981, two months after burning, and again in March and June. On each slope 20, 2 m x 4 m plots were placed contiguously in a belt running vertically down the slope from a randomly selected point. These plots were marked and returned to on each sampling date. However, due to massive increases in herbaceous and shrub seedlings at later dates, only a subset of each plot was sampled (20, 2 m x 2 m plots). Within each plot, height was measured for all plants and the ground surface covered by each individual was estimated as follows. In most instances area cover approximated a circle and thus the diameter was recorded. In other cases rectangular shapes were more appropriate and thus length and width were recorded. For all species, both herbaceous and woody, seedlings and resprouts were noted and recorded separately. Density and total coverage were calculated as described by Cox (1980). Average coverage and height for shrub resprouts, shrub seedlings, and herbs were compared within a site with a oneway analysis of variance and Fisher's Least Significant Difference Test. RESULTS-Figure 1 shows changes in density of herbs and shrubs at all sites through the first post-fire year. Few herbs were present in January and peak density was in March at the lower elevation sites, where most aboveground herbaceous material was dead by June, although at the higher elevations herb denisty remained high into June. Throughout this first year, herb density was many times greater than shrub density except at the highest elevation (site V). This herb flora was composed largely of annuals although at several sites herbaceous perennials and suffrutescents were