Page Banner

United States Department of Agriculture

Agricultural Research Service


Location: Exotic and Invasive Weeds Research

2013 Annual Report

1a.Objectives (from AD-416):
1) Assess the marsh-wide abundance, distribution and habitat/community characteristics of invasive Lepidium latifolium and endangered Cordylanthus mollis ssp. mollis, before and after implementation of weed management control strategies. 2) Assess weed treatment efficacy, plant community succession, and non-target effects across environmental gradients within the marsh. 3) In pilot scale experiments, test alternative approaches for weed control in sensitive areas where invasive weeds directly interact with vulnerable endangered plants. 4) Conduct demographic-based research to evaluate weed response to management actions, and to diagnose causes of poor population performance of native plants, and provide biologically based data for integrated weed management strategies and endangered plant recovery.

1b.Approach (from AD-416):
• GPS technology will be used to record geospatial data on population size, distribution, and habitat conditions of target weeds and endangered plants at Southampton Bay Natural Preserve, Benicia, CA. Random control and treatment plots will be stratified within by discrete within-marsh hydrogeomorphic settings that correlate with observed phenological differences in Lepidium latifolium growth. Baseline evaluation will include evaluation of L. latifolium stem density, biomass and plant community associates, and sediment physico-chemical characteristics. Post-treatment evaluations will track changes in weed density, biomass and community succession to evaluate efficacy of management actions, and the need for adaptive change in approach. A life table response experiment (LTRE) will be conducted to evaluate weed demography and the direct response of soft bird’s-beak to Lepidium alternative weed control approaches. C. mollis ssp. mollis individuals in 0.25m-2 plots within Lepidium invaded subpopulations (treatment plots) will be tagged, and survivorship and life stage transitions will be recorded. At the end of the annual life cycle, fecundity of C. mollis will be quantified. Management option impacts on C. mollis population growth rates and survivorship will be compared. Results will be directly applicable to weed control strategies, conservation management, ecological restoration and endangered plant recovery efforts. Documents Reimbursable with DOi- Fish & Wildlife. Log 39319.

3.Progress Report:

The agreement was established in support of Objective 1 of the in-house project, the goal being to develop effective weed control efforts that facilitate the recovery of Chloropyron molle ssp. molle, (soft bird’s-beak, x-Cordylanthus mollis), a federally-listed endangered tidal marsh plant, and the native plant community that is directly impacted by Lepidium latifolium (perennial pepperweed). ARS scientists used GPS-based technology to annually assess the marshwide distribution and abundance of L. latifolium and C. molle to prioritize weed treatments, provide information for adaptive management decisions, and to establish rare plant protection zones for stratified management approaches. Data were also stratified by 3 study zones across a tidal inundation gradient within the marsh, and by occupied microhabitat types. Within-patch weed abundance was quantified using 4 cover classes, and log abundance classes were used for C. molle. We detected and mapped 16.08 hectares of Lepidium latifolium in the pretreatment year (2009) with a bimodal distribution (65% along channels, 35% in marsh-terrestrial ecotones). Ninety-eight percent of the L. latifolium stands were classified as medium (31-60%) or high (61-100%) cover class. In this baseline year, we also mapped 164 subpopulation patches of the rare plant that occupied 0.49 hectares. Prior to weed treatments actions, C. molle was most abundant in un-invaded marsh plain habitat. Following four years of herbicide treatment, the area occupied by L. latifolium has been reduced by 83%, and 94% of the remaining stands are at low to trace density levels. The population size and distribution of C. molle increased in each year of the project. By 2012, we documented 546 sub-population patches of C. molle occupying an area of 1.5 hectares. The occupied area more than tripled simultaneous to herbicide management for L. latifolium control and increased by 212% compared to the pretreatment year. To improve weed management success, ARS researchers tracked the phenology and growth of L. latifolium relative to the timing of herbicide applications, evaluated the efficacy of backpack-spray herbicide applications and resprouting capacity of the weed, evaluated efficacy of hand-wick herbicide application methods within 1-m of endangered plants, non-target plant community response, and the efficacy of hand-weeding for L. latifolium control in sensitive black rail (marsh bird) nesting habitat. L. latifolium aboveground biomass and stem density were greatly reduced by backpack-spay applications of glyphosate herbicide. Independently, protected GLM analyses showed that herbicide application reduced both biomass and stem density over the study period. Below ground reserves of non-structural carbohydrates that fuel plant resprouting were also reduced in response to herbicide treatment. We calculated treatment effect sizes to illustrate the practical outcome of the herbicide treatments (variation in effectiveness) by marsh study areas and invaded microhabitats. After the first year of treatment analyses suggested the variation in suppression of L. latifolium biomass was greatest in the mid marsh study area along channels where treatment outcome was the least effective, and project managers adjusted their approach for improved efficacy. Hand-wick applications of glyphosate to L. latifolium within 1-m of any rare plant was highly effective in killing the weed, with no observed mortality to rare plants. There was an average decrease of 396 g/m2 in live L. latifolium biomass in response to manual removal in the sensitive black rail habitat areas after rail breeding season, though L. latifolium cover class did not change markedly in these areas and an increase in overall spread of the weed was observed. This is likely because seed had dispersed prior to permitted entry in the area for hand-weeding.

Careful and precise applications of herbicides by the weed management team to control L. latifolium have changed plant species composition and relative abundance of native and exotic plant species in the community in efficacy monitoring plots. The short term response to herbicide application increased the cover of other exotic species (particularly the exotic herb Atriplex prostrata; and decreased the cover of native species. However, as succession proceeded the cover of native species increased steadily each year. Short-term observations are reported here, but community succession needs to be evaluated over a longer time period. Positive responses by non-target native plant species were recorded. For example, native Senecio hydrophilus (marsh butterweed) co-occurs channel-side with L. latifolium. Following treatment of L. latifolium, S. hydrophilus has increased in abundance and showed no herbicide damage. In the pretreatment year, a single plant of sensitive Lathyrus jepsonii var. jepsonii (Delta tule pea) was observed. Two years after herbicide treatments, we observed a marked increase of L. j. var. jepsonii, to the point where it is now common along tidal sloughs. Other native plant species were observed to respond positively to L. latifolium suppression, particularly in plants that track slough channels. Notable were increases in stands of native Pluchea odorata (salt marsh fleabane), Achillea millefolium (yarrow), Grindelia stricta var. angustifolia (marsh gumplant), Euthamia occidentalis (western goldenrod), and Symphyotrichum subulatum var. parviflorum (slender aster).

Demographic evaluation through an experimental framework was utilized to assess the response of endangered C. molle to weed management practices. Upon emergence of seedlings in spring 2010, experimental plots were established within naturally occurring populations of C. molle to compare 3 treatments arranged in a factorial design with demographic variables (survival, fitness, population vital rates) of C. molle as the response variables. We established the large (~ 100 m2), replicated demographic monitoring plots in two L. latifolium-occupied microhabitats (marsh-terrestrial ecotone, channel edge) within both bayshore and mid marsh study areas reflecting inundation and salinity gradients. We tagged individual C. molle plants at seedling stage to assess seedling density, survivorship, reproductive plant density and seed rain for 72 randomly chosen subplots. In the second year of the study, endangered California clapper rails colonized in the study area for the first time in decades, entry into their potential breeding habitat was prohibited, and 50% of our experiment (all plots near channels) were abandoned. Therefore, we will not be able to parameterize a matrix model to assess population growth rates across habitat types. However, data have documented the positive effects of the weed management on the annual survival and fitness of the endangered plant population in critical marsh-terrestrial ecotone areas. In 2010 and 2011, the mortality rates of immature C. m. subsp. molle seedlings were higher where the rare plants emerged within invasive L. latifolium stands. Seedling (immature) mortality was very low among all treatments in 2012, with rates between 3 and 9%. Herbicide treatment had an immediate positive effect on demographic characteristics of C. m. subsp. molle in 2010 and 2011. In 2010, seed capsule production and fecundity were greater than any other year of this study, and greater in plots along the channel than in upland edge plots for all treatment groups. Among the most frequent potential native host plant species within C. molle demography plots were Distichlis spicata (saltgrass), Sarcocornia pacifica (pickleweed), Isolepis cernua (low club rush), Triglochin spp. (seaside arrowgrass), and Jaumea carnosa (fleshy jaumea). Exotic Atriplex prostrata (fathen) and Polypogon monspeliensis (rabbitfoot grass) were also common associates that may need to be controlled.

ARS scientists have provided scientific input to an adaptive management process that has prompted land managers to change how the project has been implemented over time to maximize the effectiveness of their effort to control Lepidium latifolium. Eradication efforts focus on stratifying the marsh by endangered species protection zones and treating the weed populations in various zones based on leaf area/weed phenology, and varying application methods to avoid impacts to endangered species. Treatment approaches have evolved every year in response to the changing character of the infestations and key species distributions, and to incorporate new methods that improve efficacy and efficiency. The project has demonstrated that careful, ecolologically-based weed management using precision applications of an otherwise broad-spectrum herbicide can be implemented in sensitive tidal wetlands for recovery of endangered species and native vegetation. U.S. Fish and Widlife Service biologists are using the project as a model for responsible weed management in endangered species habitat and tidal wetland ecosystem recovery in the San Francisco Estuary.

Last Modified: 4/17/2014
Footer Content Back to Top of Page