Soil Ecology andRestoration Group

COASTAL SAGE SCRUB RESTORATION ON

0.6 ACRE MITIGATION SITE ON

MARINE CORPS AIR STATION MIRAMAR,

SAN DIEGO, CALIFORNIA

Second Annual Report

Executive Summary
Road expansion in November 1998 on Marine Corps Air Station Miramar extended into 0.3 acres of coastal sage scrub habitat inhabited by the federally-listed threatened coastal California gnatcatcher (Polioptila californica californica). As mitigation for the disturbance, restoration of 0.6 acre of coastal sage scrub was required. On 28 September 1998, Southwest Division awarded a Cooperative Letter of Agreement, at the request of MCAS Miramar, to the Soil Ecology and Restoration Group to locate potential 0.6 acre coastal sage scrub restoration sites, prepare a document outlining these sites, and restore the site selected by the Station. In March 1999, a site adjacent to Austin Avenue was selected and approved. Soil compaction was measured and an old road on the site was found to be as compacted as a nearby actively used dirt road. To study the effects of soil compaction on container seedling survival, an experiment was installed comparing a rototilled plot and a control compacted plot. Fifty coastal sage scrub seedlings were planted in each plot for a total of 300 seedlings on the entire site. Soil samples were collected from the site and an undisturbed area and analyzed for pH, percent organic matter, texture, nitrate, total nitrogen, and phosphorus. Soil characteristics have so far been similar in the two areas except for percent organic matter and nitrates. However, it is too early in the monitoring process to determine if these differences are significant. At 76 percent, overall seedling survival was low the first year due to a late planting date and a dry La Niña year. An additional 26 seedlings were planted in March 2000 to insure the success of the project. This year, survival was 75 percent which meets success criteria and further supports the idea that the initial steep decline in survival was due to unusual La Niña conditions. Seedling survival and health in the rototilled plot has been greater than in the control plot for the duration of the project. In addition, rototilling appears to be effective as a means of weed control. A vegetation survey took place again in May of this year and results were compared to the surrounding undisturbed area. Species percent cover currently falls below the established success criteria, but this may be due, once again, to the La Niña event which would have caused seedlings to grow more slowly than in a normal year. Total species density meets success standards. Seedlings were irrigated twice a month May through September 1999, once a month May through September 2000, and will be irrigated on an as needed basis in 2001. Percent survival of container plants will be measured again in January 2001 to determine whether more seedlings need to be planted this winter. If results are acceptable, then we will hold off on planting until we monitor for species percent cover in May 2001. If results in May are still significantly lower than the required percent cover, we will plant more seedlings in the fall of 2001. Maintenance and monitoring on the site will continue through September 2002.

Introduction
Marine Corps Air Station Miramar, hereinafter referred to as the "Station", is located in San Diego County, California four miles east of the Pacific Ocean and 13 miles north of downtown San Diego (Figure 1). Road expansion work in November 1998 on Vega Road just east of Kearny Villa Road and on the east end of Ammo Road extended into a total of 0.3 acres of undisturbed coastal sage scrub habitat. This area is known to support populations of the federally threatened coastal California gnatcatcher (Polioptila californica californica). Mitigation for this project required restoration of a 0.6-acre area. In September 1998, a Cooperative Letter of Agreement was signed between Southwest Division, Naval Facilities Engineering Command and the Soil Ecology and Restoration Group (SERG) at United States International University (USIU) to locate potential 0.6 acre restoration sites, prepare a restoration plan for the selected site, and restore the site to coastal sage scrub vegetation. In November 1998, seven sites were selected for potential restoration and described in the document "Proposed Restoration Sites for 0.6 Acre Native Plant Restoration on Marine Corps Air Station Miramar, San Diego, California", dated 20 November 1998. Site number four was selected from this document and approved by MCAS Miramar, Public Works Department in March 1999 and by MCAS Miramar in March 1999.

During the first year of this project, soil strength measurements on the site showed a high degree of compaction on the dirt road bisecting the site. One-half of the road was rototilled to test the effects of soil compaction on seedling survival. In April 1999, 300 coastal sage scrub seedlings grown in the SERG greenhouse were planted on the site. Vegetation surveys and soil analyses were conducted on the site in May 1999 and compared to surrounding undisturbed areas. Maintenance in the form of irrigation, weeding, and removal of plant protection and trash occurred through September 1999.

In March of this, the second year of the project, 26 additional seedlings were planted to replace those lost during the dry summer months of 1999. Maintenance and monitoring activities continued (Tables 1 and 2) with all seedlings being irrigated once per month from May through September. Next year, the site will be watered on an as needed basis

Figure 1. Austin Avenue restoration site located on Marine Corps Air Station Miramar.

only. Vegetation monitoring and soil analysis took place again in May of this year and will continue for the next two years.

Table 1
Dates of site visits on the Station.

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

Site Description
The site is located approximately 30 meters (m) west of Austin Avenue and 100 meters northwest of the South Gate off Harris Plant Road (Figure 1). This site was selected over the other sites primarily because no future Public Works projects were scheduled that would cause another disturbance. The area requiring restoration is approximately 2.5 acres in size; however, restoration is only being conducted on 0.6 acres, an area sufficient to fulfill the mitigation requirement. Native vegetation on the site mainly consisted of Laurel sumac (Malosma laurina), Flat-top buckwheat (Eriogonum fasciculatum), Foothill needle grass (Nassella pulchra) and Tarweed (Hemizonia fasciculata). Exotic vegetation on the site included Avena and Bromus exotic grass species and Red filaree (Erodium cicutarium). Soil was mildly compacted with signs of disturbance caused by off road vehicle activity. Fauna on the site through visual sightings and presence of tracks consists of coyotes, rabbits, and rattlesnakes. The site is situated just east of San Clemente Canyon. San Clemente Canyon is composed of a coastal sage scrub community dominated by Laurel sumac, Toyon (Heteromeles arbutifolia), California sagebrush (Artemisia californica), Flat-top buckwheat and Yerba santa (Eriodictyon crassifolium). Initial disturbance appears to have been caused by the construction of a graded dirt road through the site.

Methods
Planting
On 1 March 2000, 26 additional coastal sage scrub seedlings (Table 3) were planted at the 0.6-acre site to replace those lost during the summer of 1999. These plants were grown in the greenhouse at San Diego State University (SDSU) and the shadehouse at USIU. While at the greenhouse, seedlings were initially watered three times a week and fertilized monthly. Water and fertilizer were reduced one month prior to planting and the plants were removed from under shade cloth. Seedlings were planted into holes dug with either a power auger or a shovel. Plant protection in the form of Treepees, cone shaped plastic coverings one to two feet tall with a one foot wide base, or tubex, cylindrical plastic coverings 15 to 20 cm tall, were placed over each seedling to reduce herbivory. All exotic vegetation was removed to a radius of 18 cm around each plant.

Table 3
Additional seedlings planted 1 March 2000.

Maintenance
After the additional seedlings were planted, all plants on the site were watered once a month until September when winter rains began. Plants received 1/2 gallon per watering. Water from a pump located just inside the closed Station gate next to Harris Plant Road was used to fill a 180-gallon water tank in the back of a pickup truck. Hoses were run from the tank to the site and, with the help of a small gas-powered pump, all plants were hand-watered. Originally, an irrigation system made up of 1/2 inch plastic tubing and two gallon per hour drip emitters was installed, but constant and severe damage from coyotes made this system unfeasible (Figure 2). Tubing was removed from the site, as were Treepees and tubex plant shelters from plants that had outgrown the devices.

Figure 2. A drip emitter and irrigation tubing destroyed by coyotes.

Monitoring
Documentary photos were taken throughout each stage of the project. No data was collected regarding the health and size of plants within the rototilled portion of the restoration site as compared to that of planted seedlings elsewhere, but visual observations were made. On 25 May 2000, vegetation was surveyed using two 50-meter transects on the restoration site. Transect one on the site faces east at 80 degrees from the zero point and transect two faces east at 100 degrees. Percent cover of herbs and forbs was measured using three one-meter quadrats on the site. Percent survival of container plants was also measured.

A total of six soil samples, three taken from the restoration site and three from the undisturbed area, were analyzed for total nitrogen, nitrate, available phosphorus, percent organic matter, and pH by A&L Laboratories in Modesto, California.

Results
Seedlings planted in the rototilled plot grew much larger and were much healthier than the seedlings planted on the rest of the site (Figure 3). In addition, fewer weeds grew in the rototilled section of the site (Figure 4).

Figure 3. Seedlings planted in the rototilled portion of the Austin Avenue restoration site.

Figure 4. Very few weeds grew in the rototilled portion of the restoration site.

Results from the vegetation survey conducted in May 2000 showed 9.5 percent native shrub cover and 2,040 shrubs per hectare on the restoration site (Table 4). For the first year after planting, the success criteria state that percent cover of native species should be 20 percent and species density should be 700 plants per hectare. Percent cover at this site falls short of this requirement, but species density is acceptable. Percent cover of herbs and forbs is 25 percent in comparison to the previously estimated nine percent in surrounding undisturbed areas.

Table 4
Results of vegetation survey on the restoration site in May 2000.

Overall plant survival in May 2000 was 75 percent (Table 5). The success criteria for this site require 70 percent survival after the first year.

Table 5
Percent survival of container plant species in June and September 1999 and May 2000.

Species Common name Percent survival June '99 Sept '99 May '00 Artemisia californica California sagebrush 100 94 86 Eriogonum fasciculatum Flat-top buckwheat 70 59 63 Heteromeles arbutifolia Toyon 52 46 52 Lotus scoparius Deerweed 42 52 52 Malosma laurina Laurel sumac 84 74 76 Mimulus aurantiacus Monkey flower 100 97 100 Nassella lepida Foothill needle grass 100 100 88 Rhus integrifolia Lemonadeberry 100 100 100 Salvia mellifera Black sage 98 80 75 Overall 88 76 75

Results of the chemical analyses of soils collected from the restoration site and from undisturbed areas in June 1999 and May 2000 are shown in Figure 5. Total nitrogen increased at both sites, but it is still much lower at the disturbed site than at the undisturbed site. Available phosphorus and pH are the same at both sites while the level of nitrates and percent organic matter increased at the disturbed site.

Figure 5. Soil analysis results of samples collected on Austin Ave. restoration site and surrounding undisturbed area.

Discussion
Rototilling compacted areas before planting appears to greatly increase survival and health of container seedlings. The seedlings in the rototilled section of the site stayed greener longer into the dry season and outgrew their protective shelters much faster than seedlings on the rest of the site. Decompacting the soil eases growth of plant roots, enabling them to reach groundwater much more efficiently. In addition, rototilling produced a result different from its original purpose. It also seems to aid in the suppression of weeds.

One year after planting, vegetation cover on the site was at 9.5 percent in comparison to the undisturbed area at 120 percent (greater than 100 percent cover resulted from species overlap at survey points). This figure also falls short of the success criterion of 20 percent cover. We expect a percent cover that is lower than the undisturbed site because seedlings are initially very small, but a result that falls below the required 20 percent is likely due to something else such as the unusual La Niña occurrence during the 1999 rainy season. Low rainfall last year slowed seedling growth, so they are not as large as they would have been after a normal rainfall year.

Although the vegetation density on the site and the surrounding area were similar and site density meets the success criterion, vegetation diversity was different on the two sites with more grasses and forbs on the restoration site and larger, well-established shrubs in the undisturbed area. Over time, as shrubs grow larger, vegetation diversity on the site will become more similar to that of the undisturbed area. Percent cover and density results from the survey in the undisturbed area will be used for future surveys on the site for comparative analysis. Container plant survival fell sharply during the first year, probably because of the below average precipitation during La Niña. However, it now appears to have leveled off.

The larger amount of litter due to the site's high percent cover and density of native shrubs may explain higher levels of nitrogen in the soil at the undisturbed site. The sudden change in nitrate and percent organic matter levels at the disturbed site may simply be a result of sampling technique since soil samples are not collected from the same spot each time and soil characteristics differ greatly over very short distances. It is therefore too early to tell whether these differences are significant, so we will wait until results from the next sampling period (in May 2001) to make any determinations.

Conclusions and Recommendations
Rototilling produced significant results in terms of weed control and improvements in plant health. In the future, if weeds on site become a significant problem or if it becomes necessary to plant more container seedlings because of low container plant survival or percent cover, we will rototill more portions of the site. We will also begin to take more data on percent survival of seedlings within the rototilled plot and on the rest of the site in order to continue to compare the effectiveness of this technique.

Because of the low native species percent cover measured this year, it is likely that we will plant more seedlings this fall. In January of 2001 we will collect percent survival data and if the results have fallen significantly since May, we will plant. However, if percent survival is still at an acceptable level, we will wait until May of 2001 and the results of that year's vegetation survey to see if percent cover is low enough to require the planting of more seedlings.

Maintenance on this site will continue with irrigation on an as needed basis. If more seedlings need to be planted, these will be watered twice each month for the first summer after planting and once each month for the summer after that. Treepees and tubex will be removed as plants outgrow the shelters and any remaining pieces of tubing from the irrigation system will be removed. Monitoring of both soil and vegetation will continue in May each year through 2002.

First Annual Report (January 13, 2001)

Final Report (February 17, 2003)