Primarily, we desire to underscore the need for thoughtful and strategic restoration that is paired with invasive species control. In a cautionary example of the importance of such a pairing, managers in southwestern riparian ecosystems achieved success in controlling the invasive Salt Cedar , only to discover that without combining control efforts with native vegetation restoration, the endangered Southwestern Willow Fly Catcher struggled to locate nesting habitat . There are many impediments to effective adaptive management in the Delta, including a general trend of managers’ failure to evaluate and synthesize management results . However, the present case study is an example of state and federal agency employees collaborating with academic researchers to address this common pitfall by evaluating the results of management operations and incorporating the evidence into decision-making about future management activities. More broadly, we hope this case study is a useful example for managers around the world who manage invasive species in the context of aquatic food webs. Modern ecological restoration projects generally focus on short term interventions due to limited funding, finite resources, dry racking and short policy or grant cycles. We define short-term intervention efforts as the “implementation phase” recognized by the Society for Ecological Restoration, which includes the initial 1–5 years of restoration .
This implementation phase involves substantial money, labor, equipment, and other resources to alter the abiotic environment, remove exotic species, and introduce native species. The implementation phase initiates ecosystem recovery by targeting and manipulating key determinants of successional pathways . After short-term restoration efforts cease, the restored ecosystems become subject to ambient drivers of succession, such as the natural recruitment of plants via existing populations and uncontrolled environmental conditions that favor some species over others . For desired native populations to persist past the initial implementation phase, natural recruitment and environmental conditions must favor these native species. Landscape context has long been recognized as an important factor influencing the trajectory of a restoration site . Landscape heterogeneity, such as grasslands scattered with trees, can contain species-rich microhabitats that increase overall species diversity . Land use history also affects species diversity, as undisturbed landscapes can harbor species-rich seed banks. Seed availability and dispersal also affect species diversity in restoration sites, and there is a tendency for restoration sites to become dominated by weedy species that are already present at the site . For natural recruitment into a restored site to be dominated by desirable species, the most abundant populations in the matrix surrounding the restoration site should be native species . Initial restoration plantings often establish small populations of desirable species that frequently exist as patches within a fragmented landscape otherwise dominated by undesirable exotic species and isolated from other native populations .
In such settings, after the implementation phase ends, isolated restored sites may become reinvaded quickly by undesirable species from the surrounding landscape. Even if there are native individuals present, environmental conditions may prevent population growth and even result in local extirpation . For example, restored wetlands that established an average of 28 native species within the implementation phase subsequently experienced a decline in richness to 12 native species 6 years later . Reinvasion of restored ecosystems by exotic species is a common challenge faced by restoration practitioners, and it is well known that exotic species are particularly adept at colonizing open niche spaces following disturbances and under shifting climatic conditions . One-time exotic species removal efforts can also lead to a secondary invasion, wherein another invasive species establishes after the removal of the original invasive species . Pearson et al. found that secondary invasion of exotic species occurred in all 60 of the weed management projects they surveyed in a global meta-analysis. They found a strong inverse relationship between secondary invader abundance and original invader abundance, suggesting that secondary invaders took advantage of reduced competition and more resources after the original invaders were eradicated. Vernal pool assemblages in California, U.S.A., are especially susceptible to reinvasion by exotic plants after initial restoration, particularly by annual grasses from Europe . In California’s Mediterranean climate, vernal pools form atop an impermeable subsurface soil layer during the cool, wet winters and then dry out during the warm, dry summers . Endemic plant species flourish in this unique environment with adaptations that allow them to survive prolonged flooding, while also growing and reproducing quickly before pools completely desiccate during the summer . Specialist species that have adapted to withstand this hydrologic regime can take advantage of the lower amount of competition in these harsh environments .
Some native plant species, such as Lasthenia fremontii , are only found in the deepest, most inundated zone of the pool and cannot withstand drier conditions, whereas other species, such as Limnanthes alba , are adapted to slightly drier conditions along the shallower edge zones of the pool and cannot withstand extreme flooding events . The pool landscape can be heterogeneous within the space of a few meters, which has direct implications for native species growth and persistence and thus restoration and management. In addition, pools typically exist within a landscape matrix that is dominated by invasive exotic plant species, resulting in edge effects wherein the pool margins are exposed to the invasion front of surrounding exotic species . Vernal pool restoration projects have had varying levels of success, particularly in southern California . This may be due to variable site characteristics and competitive pressures from exotic plant species in some zones of the pool or some parts of the pool complex . Restoration actions often consist of topographic excavation of deeper pool basins, resulting in prolonged flooding of the central zone of the pool, followed by the addition of native plants. Creating wet abiotic conditions allows any added native seed to grow and reproduce without competition from invasive exotic species that cannot withstand inundation. Yet, cannabis curing as elevation increases up to the pool’s edge, conditions become drier, and the community is more susceptible to invasion by generalist European grasses that can opportunistically invade drier open niche space . Gerhardt and Collinge showed that, even when native species were abundant, a longer inundation period was needed to preclude subsequent exotic invasion. They manipulated the inundation period in a greenhouse experiment and found that, although the growth and reproduction of some exotic species were reduced when grown with native species, longer inundation significantly decreased the survival of exotic species. A field study by Faist and Beals similarly found that pools with higher invasive species cover also had shorter inundation periods. Drier years can cause an increase in exotic forbs in pool basins, likely due to the lack of abiotic resistance normally afforded by flooding . In addition to the abiotic conditions that need to be established in the pool center to reduce its invasibility, biotic manipulation of the pool edges may need to be a continual effort to prevent exotic reinvasion . Marty reported that an increase in exotic species cover coincided with the discontinuation of a vernal pool site’s weed management program, which had included grazing. Marty found that reintroducing grazing allowed pool plant communities to recover significantly higher native cover than ungrazed pools, with the greatest increase in native plant cover found around the pool edges. These edge effects, or conditions at the edges of sites that alter abiotic conditions, species composition, and ecological processes, can often have detrimental ecological consequences . For example, hotter, drier, and more variable conditions along exposed forest edges can result in higher tree mortality rates, and exotic propagule pressure and anthropogenic disturbance can correlate with higher exotic plant species and lower native plant species on the edges of preserved grasslands . In vernal pools, pools with more edge area exposed to surrounding unrestored grassland may also be more susceptible to similar edge effects, including invasion. Habitat fragmentation studies have shown that fragments with higher perimeter-to-area ratios exhibit higher exotic cover .
Restoration efforts in these drier zones often include weeding out invading exotics, which is generally not needed in the central zone where inundation excludes invasive species . This biotic manipulation can allow native species to reestablish, but the duration of weeding is often limited to the implementation phase due to financial constraints . Overall, the management challenges faced by restored vernal pool assemblages are tenacious and long-lasting, while most restoration projects are restricted to the short timescale of the implementation phase. To date, most research on vernal pool restoration has been focused on short-term measures of restoration success, but it is unclear how successful short-term interventions are in the long run. First, we conducted a 3-year study on a complex of pools that were transitioning from the implementation phase to the post-restoration phase during the study period. By evaluating changes in vegetation composition in these pools during this pivotal transition period, we asked: How did exotic plant abundance and richness change in these restored pools over time? How did native plant abundance and richness change in these restored pools over time? As restored pools receive less weeding and native planting over time, we might expect the reinvasion of exotic grasses from the surrounding grassland matrix. Our second approach involved a broad survey of 69 vernal pools from nine different restoration projects carried out over 33 years, which allowed us to explore how climatic and landscape conditions correlate with the abundance and diversity of plant species within restored vernal pools after the implementation phase. If exotic plant species reinvade vernal pools over time, we might expect various site landscape factors to influence the plant assemblages. For example, pools that experience more precipitation and/or have deeper basins may sustain longer inundation periods that favor more native species, while pools that have more edge area exposed to the exotic-dominated surrounding grassland may be more susceptible to invasion . We asked: What abiotic factors correlate with higher exotic plant abundance and richness in restored pools over time? What abiotic factors correlate with lower native plant abundance and richness in restored pools over time?We studied restored vernal pools on land managed by the University of California, Santa Barbara , the Isla Vista Recreation and Parks District, and the City of Goleta, in Santa Barbara County, California, U.S.A. . This land is part of unceded ancestral territory of the Chumash people. The study areas lie within 1 mile of the Pacific Ocean and experience a Mediterranean climate with cool and wet conditions from November to April and warm and dry conditions during the remainder of the year . Rainfall averages approximately 43.18 cm per year with high variation associated with extreme rainfall events and droughts. The proximity of the area to the Pacific Ocean moderates winter lows, and frost is rare. Summer fog moderates summer highs, although offshore “sundowner” winds may bring hot dry conditions to the area, especially in the late summer and fall . Soil formation is dominated by weathering of uplifted shales, and soils have a high clay content. Soils are Mollisols, with the dominant soil series being Concepcion fine sandy loam and Diablo clay .We monitored 7 restored vernal pools within UCSB’s North Parcel, which consists of vernal pools built amidst university faculty housing. The pools were created between 2011 and 2014 by grading to form pool basins ranging from 67 to 425 m2 in area and 14–18 cm deep . Approximately 70 species of locally-sourced native plants were introduced to pool basins, including species endemic to vernal pools and generalist wetland and upland species. Most species were introduced by planting seedlings in patches to mimic landscape patterns generally observed in nature, according to soil types, hydrology, and other site factors. Installed plantings were watered using movable drip irrigation and hand-watering until establishment was achieved. Some annual species were direct-seeded. Exotic species were mainly controlled by hand-weeding, although solarization, herbicide, and green flaming treatments were also employed to a lesser degree. All these restoration actions took place within a 5-year implementation phase. Within each restored vernal pool, we established a series of permanent monitoring quadrats. We delineated each pool into central , transition , and upland zones . Within each of these zones, we haphazardly placed three 1 m2 quadrats, for a total of 9 quadrats per pool. We monitored the vernal pools monthly from November 2016 to December 2019. Because the pools were different ages at the start of the experiment, sampling over 3 years allowed us to evaluate the vegetation community in pools ages 2–9 years old.