Soil Ecology and Research Group last update February 18, 2003

COASTAL SAGE SCRUB RESTORATION FOR
GNATCATHER MITIGATION ON WEAPONS
SUPPORT FACILITY SEAL BEACH, FALLBROOK
DETACHMENT

Third Annual Report

Abstract
Weapons Support Facility Seal Beach, Fallbrook Detachment is located in northern San Diego County, bordered to the east by Fallbrook and to the west by Camp Pendleton Marine Corps Base. As mitigation for the construction of a paved ordinance truck class holding yard that caused removal of 0.6 acres of coastal sage scrub inhabited by the coastal California gnatcatcher, the Navy was required to restore disturbed coastal sage scrub habitat at a ratio of 2:1. In September 1998, the United States Navy signed a Letter of Agreement with the Soil Ecology and Restoration Group (SERG) at San Diego State University (SDSU) to perform native plant restoration research on a 1.2 acre area of disturbed coastal sage scrub adjacent to the holding yard construction area. Initial work on the site began with the installation of experimental topsoil plots using soil removed from the construction area. Soil was placed in two 50 m by 50 m plots on the restoration site while two 50 m by 50 m plots remained bare as a control. In order to protect the site from grazing cattle, a 300 m barbed wire perimeter fence was constructed in February 1999. In February 2000, 478 coastal sage scrub seedlings grown in the SERG greenhouse at SDSU were outplanted. Only a few of these seedlings were planted in the plots covered by the salvaged topsoil because numerous volunteer seedlings had already germinated. Transplants were irrigated twice each month throughout the summer. In March 2001, 34 additional seedlings that were not ready in February 2000 were transplanted to the site. All seedlings were irrigated during the summer 2001 and will not require supplemental water for the remainder of the project. Soil analyses and vegetation surveys were conducted in Spring 2000 and 2001 and results were compared to baseline data collected in 1999. Soil nutrient levels as well as fungal hyphae and bacteria numbers currently appear normal. The restoration site meets the success criteria for native shrub species density, but falls slightly below the criteria for native shrub percent cover and percent container survival. However, the numerous volunteer native shrubs that have sprouted from the salvaged topsoil applied to the site make it unnecessary to install additional transplants. Surveys comparing species density in the experimental topsoil plots and the bare plots showed that the 400 seedlings per acre that are usually planted for a coastal sage scrub restoration site can be decreased in areas covered with salvaged topsoil because it appears that volunteer seedlings significantly increase density. The addition of tilled topsoil also appears to be beneficial to weed species establishment, so extra effort is required for weed control. During the 2001/2002 rainy season, work has focused on spraying exotics with RoundUp Pro herbicide. Soil and vegetation monitoring will be conducted in April 2002.

Introduction
Weapons Support Facility Seal Beach, Fallbrook Detachment, located in northern San Diego County, is bordered to the west by Camp Pendleton Marine Corps Base and to the east by the city of Fallbrook. This 7,800-acre facility is used for weapons storage by the United States Navy and Marine Corps. In November 1998, construction began on a paved ordinance truck class holding yard located in the northeast section of the facility. The construction of this holding yard required the removal of 0.6 acres of coastal sage scrub vegetation that was inhabited by the endangered coastal California gnatcatcher (Polioptila californica californica). As mitigation for this loss of valuable habitat, enhancement or creation of new coastal sage scrub habitat at a ratio of 2:1 was required by the draft Programmatic Uplands Biological Assessment for neighboring Camp Pendleton Marine Corps Base and by the United States Fish and Wildlife Service. In compliance, the Department of the United States Navy signed a Letter of Agreement on 30 September 1998 with the Soil Ecology and Restoration Group (SERG) at San Diego State University (SDSU) which states that SERG will restore the required 1.2 acres while conducting research on native plant restoration techniques. Research and restoration will continue through September 2003.

During the first year of the project, efforts focused on creating experimental topsoil plots. Topsoil excavated from the holding yard construction site was transported to the restoration site on 30 November 1998 and placed in two 50 m by 50 m plots to a depth of 25 to 40 cm. Vegetation removed from the holding yard site was also added to the plots and crushed with a tractor blade. Erosion control was not installed because there were no signs of sediment loss during the 1998/1999 winter rainy season. Because of the presence of grazing cattle in and around the site, a 300 m barbed wire perimeter fence was constructed on 4 February 1999 to prevent cattle from entering the area and destroying restoration efforts. Monitoring of the plots, including chemical soil analysis, soil bacterial numbers, soil fungal biomass, and vegetation surveys, began in May 1999 and will continue until 2003. Seed collected at Fallbrook beginning in October 1998 was used to grow coastal sage scrub seedlings in the greenhouse at SDSU. In February 2000, 478 container seedlings were planted on the site. Seedlings were watered throughout the summer months and the site was hand-weeded. Site monitoring revealed that the addition of topsoil salvaged from coastal sage scrub habitat and applied to a restoration site dramatically increases the rate of native shrub establishment and reduces the need for container transplants.

In March 2001 of this, the third year of the project, 34 Baccharis pilularis seedlings were planted that were not ready in February 2000. Site maintenance including weed control, soil analysis, and vegetation monitoring continued as it will through 2003. See Tables 1 and 2 for dates of site visits and personnel involved.

Table 1
Dates of site visits on the Station.

Table 2
Names of personnel involved in the field and the greenhouse.

Site Description
The 1.2 acre study area is located east of Copperhead Road, directly across from the ordinance truck class holding yard, and is sloped approximately 8 degrees with a southeast aspect (Figures 1 and 2). The area was selected as the mitigation site due to its close proximity to the disturbance site and to undisturbed coastal sage scrub inhabited by the California coastal gnatcatcher, its near lack of native vegetation, and its ease of vehicle access. Existing vegetation on the site consisted of a few Artemisia californica (California sagebrush), Nassella pulchra (Purple needlegrass), Eremocarpus setigerus (Dove weed), and non-native species such as Bromus madritensis (Red brome) and Erodium botrys (Red-stem filaree). Before restoration began, fauna visually observed on the site included rabbits and squirrels. Disturbance to this site and the surrounding area was originally due to the installation of a now abandoned railroad to the east of the site. The use of the area for cattle grazing during the spring each year continued to impact the site until it was fenced in on 4 February 1999.

Methods
Maintenance and Planting
Seedlings that were planted in February 2000 and watered during the summer following planting, were not watered from October 2000 through May 2001 because winter rains were sufficient for plant survival and growth. The species list in the restoration plan for this project originally included 22 Baccharis sarothroides seedlings to be transplanted onto the site. These individuals were not ready at the time of planting in February 2000, so their transplanting was delayed. In the meantime, a second survey of the surrounding undisturbed vegetation revealed that what was thought to be B. sarothroides was actually B. pilularis. Therefore, in March 2001, 34 B. pilularis individuals were transplanted to the site. Planting holes were dug by hand, pre-watered, and plants were immediately installed and post-watered. Basins approximately two feet in diameter were constructed around each plant as a means of conserving water during irrigation. Each plant was given

(Figure 1)
Figure 1. Coastal sage scrub restoration site located on Weapons Support Facility Seal Beach, Fallbrook Detachment.

(Figure 2)
Figure 2. Aerial photo of restoration site.

a Tubex plant shelter to aid in reducing herbivory and desiccation. The 34 new B. pilularis seedlings were watered twice each month through June 2001. The following month, both the new and the original transplanted seedlings were watered once each month until October 2001 when winter rains took over. Plants received approximately 1/2 gallon per watering. Basins were filled by hoses hooked up to a 180-gallon water tank in the back of a pickup truck.

Due to scheduling problems, a licensed pesticide applicator was not able to spray the weeds on site during the 2000-2001 rainy season. Instead, weeds were cut in April and May 2001 with a gas-powered weed trimmer (Figure 3). Weeds (mostly Brassica nigra and Erodium botrys) were particularly tall and dense in the topsoil addition plots, so before cutting began, native plants were flagged and a buffer zone was hand-weeded around them (Figure 4). The cut weeds were left on site as a mulch layer to reduce the number of weeds that germinate during the 2001-2001 rainy season. Treepees and Tubex shelters were removed from seedlings as they outgrew their protective devices.

Figure 3. Cutting weeds with a gas-powered trimmer.

Figure 4. Hand-weeding around a transplant and an Artemisia californica volunteer.

Monitoring
In March 2001, all surviving transplants were counted and percent survival was calculated.

In April 2001, a total of ten soil samples were collected on-site; four from the two topsoil-addition plots, four from the two bare (control) plots, and two from the surrounding undisturbed coastal sage scrub. All samples were sent to A&L Western Agricultural Laboratories and analyzed for phosphorus, pH, nitrate, total nitrogen, and percent organic matter. The samples were also analyzed for fungal hyphae length and bacterial content at San Diego State University. Over an extended period of time, a relative comparison of activity, both fungal and bacterial, allows us to determine if the below ground activity occurring in the restoration site is moving toward or away from the levels found in the undisturbed reference site. Fungal hyphae length is determined using a hyphal length extraction technique (Hanssen et al., 1974; Schuepp et al., 1987; Bardgett, 1991; Frey and Ellis, 1997). Bacterial numbers are determined using the europium staining method (Anderson and Westmoreland, 1971; Anderson and Singer, 1975; Trent, 1992; Conners et al., 1994) that stains only the DNA of living cells.

Vegetation on the restoration site was surveyed on 13 June 2001 according to a revised version of the California Native Plant Society’s (CNPS) field sampling protocol. Percent shrub cover was measured using two 100-meter transects with points every one meter on the restoration site. Percent cover of herbs and grasses was measured using three one-meter square quadrats on the restoration site. Shrub species density was measured by counting shrubs within a 2.5-meter wide corridor on both sides of each transect. A comparison of the vegetation in the topsoil plots and the bare (control) plots was made using three one-meter square quadrats in each of the plots for a total of 12 quadrats. Documentary photos of the site and transects were taken during the summer 2000.

Results
Percent survival of transplanted seedlings was 74 percent (Table 3), six percent less than the 80 percent survival required by the success criteria for 2001.

Table 3
Container plants originally planted and surviving one year later.

This year, phosphorus, pH, nitrate, total nitrogen, and percent organic matter were very similar at all three sample locations (Figure 5).

Figure 5. Results for soil analyses of samples collected in December 1998 and April 1999, 2000, and 2001.

This year's hyphal length extraction results (Figure 6) revealed that the amounts of fungal hyphae found in both the topsoil and bare (control) plots is very similar to that found in the undisturbed reference site. The europium staining results for bacteria (Figure 7) this year were also very similar among all three sample locations.

Figure 6. Fungal hyphae length in soil samples taken from topsoil addition and bare (control) plots at the restoration site and one undisturbed area.

Figure 7. Bacterial numbers in soil samples taken from topsoil addition and bare (control) plots at the restoration site and one undisturbed area.

Native vegetation cover on the restoration site this year consisted of 17 percent shrubs and 27.2 percent herbs and grasses (Table 4). Shrub cover is three percent lower than the required 20 percent listed in the success criteria for 2001 and 6.5 percent lower than the shrub cover at the undisturbed site. Cover of native herbs in the restoration site is 14.2 percent higher than in the undisturbed site.

Table 4
Percent cover of native shrubs and herbs on the restoration site.

Native shrub density (Table 5) on the restoration site is 4,100 plants per hectare, much higher than the 1,100 plants per hectare found in the undisturbed reference area and higher than the 800 plants per hectare required by the success criteria for 2001.

Table 5
Density of native species on the restoration site.

The plants comprising herb and grass cover in the two topsoil addition plots are all weed species (Table 6). Over 87 percent of the weeds are E. botrys. The bare (control) sections of the restoration site also contain a high percentage of E. botrys (50.8 percent), but native species account for 27.8 percent of total cover.

Table 6
Comparison of relative percent cover of herbs and grasses in the topsoil plots and bare (control) plots.

Discussion
Work in 2000/2001 focused on planting the B. pilularis seedlings that were not ready at the time of the original planting in February 2000, irrigating all transplants throughout the dry summer months, and continuing to collect data on transplant success, soil health, and the potential benefits of using salvaged topsoil to restore coastal sage scrub habitat.

Percent survival of transplanted seedlings fell below the required success criteria for the 2001 monitoring season. However, additional seedlings were not planted because the calculated 74 percent survival does not include the hundreds of volunteer Eriogonum fasciculatum, Artemisia californica, and Lotus scoparius individuals that have sprouted from the topsoil-addition plots. If these were added to the total, survival would be well over 80 percent.

Results of soil analyses were used to determine whether levels in the restoration plots are, as a sign of recovery, beginning to follow the same trends as those in the undisturbed reference site. In 2000, phosphorus and nitrate levels in the topsoil plots were dramatically higher than in the bare and undisturbed soils. Likewise, total nitrogen in undisturbed soils peaked and percent organic matter was high in the bare and undisturbed soils. Contrarily, 2001 results revealed a strong similarity among soil nutrient levels at all three locations. Not only were nutrient levels similar among locations, but also to the results from 1998 and 1999. This indicates that the 2000 results were likely not indicative of a trend in soil health, but rather represent a localized anomaly. Therefore, there is no need for soil remediation for these parameters at this time. Similarly, soil pH results do not indicate a need for remediation since readings have remained constant over time.

It appears that fungal activity at the restoration site, for both topsoil and control plots, is currently similar to that found at the undisturbed reference site. This similarity was expected for the topsoil plots but not for the bare (control) plots. This same premise also applies to the bacteria results where low bacteria counts were expected in the bare plots but not in the topsoil plots. Below ground activity, however, is extremely variable, both temporally and spatially, and for this reason individual sample results are not as important as the trend that appears over several years of sampling. However, since the results of 2000’s chemical analyses revealed that last year’s samples likely represent a localized anomaly, it is difficult to compare this year’s fungal and bacterial data to last year’s. In addition, since no fungal or bacterial data was collected for the undisturbed site in 1999, it is difficult to determine the normal trend in fungal activity and whether or not the two sections of the restoration site are following it. Hopefully, soil samples collected and analyzed in 2002 and 2003 can be compared to 2001 data and a trend will be revealed. In the meantime, the increase in fungal and bacterial levels between 1999 and 2001 in the two sections of the restoration site indicates a general improvement in soil health.

The percent cover of shrubs at the restoration site fell slightly below the required success criterion for 2001, but additional transplants will not be installed. There are two primary reasons for this. First, shrub cover is low by only three percent. Second, the percent cover of native herbs on site is more than double that found in the undisturbed site. This is an indication that the site is healthy enough to successfully support a large number of natives. The fact that they are not shrubs is no cause for concern. Rather, it illustrates the natural progression of a disturbed site as it moves from a vegetation community dominated by grasses and annuals to a mature shrub-dominated community.

Shrub density at the restoration site for the 2001 monitoring season is over five times higher than the required success criterion and four times higher than at the undisturbed site. This is likely due to the large number of A. californica, L. scoparius, and E. fasciculatum volunteers that have sprouted from the added topsoil. Clearly, these results indicate no need for additional transplants.

The comparison of relative cover of herbs and grasses between the topsoil and bare (control) plots revealed interesting results this year. The topsoil plots contained entirely exotic species while the bare plots, although they too contained a large number of exotics, supported nearly 28 percent native species. The original intent of adding topsoil to the site was to introduce a native seed bank. This worked to some extent as can be seen by the success of native shrub establishment, but it also seems to have encouraged the establishment of exotics. This may be the result of, during the process of spreading the topsoil, native seeds getting buried too deeply to allow for germination. Another potential explanation for this is the observation that seedlings appear to have an easier time establishing in looser, tilled soils (Johnson, 2002). The topsoil salvaged from the construction site was not compacted as it was applied to the restoration site and the weeds in these plots have grown very tall and lush. The surrounding soil, however, had a history of being grazed by cattle, and is therefore highly compacted; weeds here are numerous, but remain small and low to the ground. It is possible that the compaction helps prevent invasive weed species from growing too large and, in turn, preventing native species from being crowded out.

Visual observations of the site between October 2000 and September 2001 include its use by various wildlife species including a burrowing toad (Bufo sp.), a garter snake (Thamnophis sp.), and numerous roadrunners (Geococcyx californianus) and ravens (Corvus corax). The attractiveness of the site to wildlife species is a good indication of its increased recovery. The exclusion of cattle from the site also seems to be speeding its recovery. For example, within the enclosure, Nassella sp. volunteers are thriving even though the soil is highly compacted, while outside the enclosure, there are only a few, stunted individuals.

Conclusion and Recommendations
The beginning of the winter rains in fall 2001 marked the end of the irrigation schedule for the transplants. By summer 2002, the seedlings will have been in the field for over two years and should be hearty enough to withstand the summer dry season without supplemental water. The B. pilularis, although they will only be one year old by summer 2002, will also not be watered. This species is a very hearty plant and a fast colonizer of disturbed areas, and one year of irrigation should be sufficient to ensure its survival. Vegetation health and survival will be assessed again in May 2002. It is recommended that the fence surrounding the restoration site remain in place as long as cattle use the surrounding land. Keeping the cattle out appears to increase species diversity and density, a quality that will benefit native wildlife species.

Results of soil chemical analyses have so far produced no cause for concern. Soil bacterial and fungal analyses have not produced conclusive results, but what little information is available indicates that soil health is progressing. Samples will be collected again in April 2002 and 2003 and if trends indicative of poor soil health appear, corrective soil amendments will be applied.

The addition of topsoil to a restoration site for the purpose of introducing a native seed bank has worked well for shrubs in this case, but it has also increased weed establishment. Special attention should be paid to weed control when tilling native topsoil or introducing salvaged soil from another location. Weed control will begin during the 2001/2002 rainy season when weeds are approximately four inches tall and will continue until the rains stop.

First Annual Report (June 3, 2002)

Second Annual Report (December 2, 2001)

Fourth Annual Report (February 18, 2003)

Final Report (March 9, 2004)