A site within San Timoteo Canyon was revisited, 13-14 years after treatment, to look at long-term effects of Arundo donax removal. The data obtained were used to determine whether A. donax had re-invaded, other non-native species had established the area, or if native species were able to grow in place of the removed A. donax. The previous removals included a combination of grinding large patches of A. donax and then foliar spraying, foliar spraying of uncut plants, and direct spraying of hand cut stems, depending upon the location and size of the plant. The effects of the A. donax removals within San Timoteo Canyon were analyzed in relation to new percent cover of the plant species, other more recent removals, and areas that did no experience removal procedures. The project included the use of data provided by the Inland Empire Resource Conservation District (IERCD) as well as the collection of data from randomized plots to generate plant species percent cover. Plant percent cover data analyzed for this paper had been collected from eight 15 by 15 foot randomly selected plots within an overall project site of 42.3 acres. Additional sites were used to investigate what can happen if A. donax is not removed from an area into which it has been introduced., the short-term effects of A. donax removal methods, and the role the ever-changing characteristics of riparian areas can play in their own restoration. These additional sites included aerial photographs supplied by IERCD of an ecologically similar area, a plot with a more recent A. donax removal date, as well as photographs and data of a site subject to natural recovery. Based on these comparisons it is concluded that the treatment methods used led to a lessened presence of A. donax, and that other invasive species did not grow in its place. Further, as the removal procedures within the project area occurred approximately 13 to 14 years prior, it can be concluded that there is no regrowth of A. donax and that many native species have been able to re-inhabit those areas previously infested by A. donax. The treatment methods used were successful without the need to continually disrupt the habitat and allowed for the habitat to recover naturally once the invasive species had been removed.

Increased fire frequency, livestock grazing and other disturbances have caused type-conversion of native chaparral to non-native annual grasslands. Competition with non-native annual grasses hinders shrub establishment, presenting challenges in restoration. This study examined the feasibility of chaparral restoration in a Mediterranean-type climate, on a type-converted landscape in San Timoteo Canyon, Riverside County, California. The objectives of this study were to (1) compare the effectiveness of the broad-spectrum herbicide, glyphosate, to that of the grass specific herbicide, fluazifop, at eliminating non-native annual grasses; (2) compare the success of seeding or planting native seedlings as a means of restoration; (3) analyze the soil seed bank to determine if there was a relict native species seed bank on site; (4) test the field application of smoke-water as a technique to manipulate the soil’s native shrub seed-bank for restoration. A factorial design was utilized, with three replicates: four restoration treatments (no restoration, smoke-water application, seeding, and planting seedlings) across three herbicide treatments (a January glyphosate application followed by an early-March fluazifop treatment, an early-March fluazifop treatment only, and a control treatment with no herbicide). Glyphosate application followed by fluazifop application proved to be more effective at reducing non-native annual grasses and benefitted shrub establishment more effectively than the fluazifop and control treatments. The application of glyphosate in winter promoted shrub establishment and growth, and increased soil moisture, while preventing the springtime increase in grass cover that was seen in the control and fluazifop-only treatments. The four chaparral species that were planted (Adenostoma fasciculatum, Eriogonum fasciculatum, Quercus berberidifolia, Rhus ovata) did not behave similarly. No Quercus berberidifolia individuals survived. Eriogonum fasciculatum did not exhibit differences in survival in any herbicide treatments, but exhibited higher live plant canopy volume in plots that received glyphosate with fluazifop follow-up treatments. Adenostoma fasciculatum and Rhus ovata showed higher survival in plots that received glyphosate with fluazifop follow-up treatments; however, neither exhibited differences in live plant canopy volume between herbicide treatments. These results persisted until the end of the study, 14 months after planting. Seed application of seven species (Adenostoma fasciculatum, Artemisia californica, Eriogonum fasciculatum, Gutierrezia sarothrae, Quercus berberidifolia, Rhus aromatica, and Rhus ovata) was not successful at this study site. Seed bank analysis revealed that there was a high density of non-native seeds present, and that the relict native shrub seed bank was insufficient for restoration on this site. Smoke-water application at the study site proved to be ineffective at manipulating the soil seed bank, as no chaparral species emerged. Results from this study suggest that winter glyphosate application followed by a fluazifop application in a Mediterranean-type climate may benefit shrub re-establishment more than spring fluazifop-only treatments. Results indicate that winter herbicide treatment will eliminate competitive non-native annual grasses early in the growing season, resulting in increased soil moisture availability for chaparral shrubs to utilize. Winter glyphosate application, supplemented with the planting of seedlings, could be a successful technique to re-establish chaparral shrubs on type-converted slopes.