| Title |
Investigators | Department | Objectives | Approach Keywords | Progress Reports | Impact Statements | Publications | |
Project * COL00676 | |
| Title | *Ecology and Management of Invasive Weeds on Colorado Rangeland |
| Investigator(s) | Beck, KG |
| Department | Bioagricultural Sciences and Pest Mgmt. |
| Objectives | 1. The overall objective of the invasive weed modeling project is to provide an understanding of the ecology of invasive, perennial weeds and their interactions with native vegetation to prioritize weed management decisions and to promote rangeland productivity. The overall objective will be achieved by addressing each of the following objectives: a. Develop a theoretical framework of interactions between invasive, perennial weeds and native vegetation, and effects of environmental factors (soil texture, precipitation, and temperature) on these interactions. b. Identify the combinations of environmental factors that make sites most susceptible to invasion and dominance by invasive, perennial weeds. c. Predict long-term invasion and expansion dynamics of invasive, perennial weeds and successional dynamics of native vegetation for sites differing in soil texture, precipitation, and temperature. 2. The overall objective of the diffuse knapweed (Centaurea diffusa) project is to improve the biocontrol of this weed by Sphenoptera jugoslavica and this objective will be achieved by addressing each of the following objectives: a. Enhance the control of diffuse knapweed by Sphenoptera jugoslavica by eliminating those rosettes that are too small to support successful oviposition by the beetle by applying low rates of picloram or clopyralid in mid-June or fall. b. Arrest the growth of diffuse knapweed rosettes without killing the plant by applying low rates of picloram or clopyralid in June thereby allowing consistent, successful penetration of petioles by the root beetle larvae regardless of weather conditions. c. Decrease dense, rosette populations to thresholds more susceptible to control with the insect by applying low rates of picloram or clopyralid in June or in fall. 3. The objective of the yellow toadflax (Linaria vulgaris) project is to provide a better understanding of the interactions of this weed with the abiotic and biotic components of the communities that it invades. The objective will be achieved by addressing each of the following objectives: a. Determine whether the occurrence of yellow toadflax is correlated with environmental factors such as habitat, slope, aspect, soil texture, soil organic matter, soil carbon:nitrogen ratios, precipitation, disturbance, or plant community composition. b. Determine whether there is significant genetic variation within and among populations of yellow toadflax in Colorado and whether the variation can be predicted by certain infested site characteristics. |
| Approach | 1. A conceptual model will be developed that describes the interactions of invasive perennial weeds with abiotic and biotic components of the rangeland environment where invasion occurs. The model will be developed from manipulative field experiments associated with the project and from data found in the literature. Model validation will be accomplished through field surveys. 2. Sublethal rates of picloram and clopyralid will be applied to diffuse knapweed in spring or fall to manipulate rosette densities such that more optimum sized rosettes for the beetle comprise the population. The spring herbicide treatments also will be used to mimic an environmentally induced growth arrest of diffuse knapweed that occurs annually in its native range; this growth arrest is necessary for successful larval penetration into diffuse knapweed's root system. Data on life cycle transitions will be collected to assess the effects of herbicide treatments on the growth arrest and composition of the diffuse knapweed populations relative to rosette size. 3. Two drainages in the Flat Tops Wilderness in the White River National Forest in western Colorado will be surveyed for yellow toadflax and compared to sites in the same drainages that are not infested with yellow toadflax. Site characteristics will be recorded and correlation analysis will reveal if yellow toadflax occurrence correlates with environmental factors such as slope, aspect, soil texture, soil organic matter, soil pH, soil carbon to nitrogen ratios, disturbance, and plant community. Genetic analysis of yellow toadflax also will be conducted. Six Yellow toadflax biotypes will be collected from three locations in Colorado and genotypic differences of nuclear DNA will be determined using Amplified Restriction Fragment Polymorphism (AFLP) technology. These same six biotypes will be compared for their susceptibility to two classical biocontrol agents and two herbicide treatments to determine whether genetics influence control. |
| Keywords | invasive weeds, rangeland weeds, noxious weeds, diffuse knapweed, centaurea diffusa, yellow toadflax, linaria vulgaris, invasive weed management, noxious weed management |
| Progress Reports | |
| 2000 | Diffuse knapweed management: Sphenoptera jugoslavica negatively influences diffuse knapweed populations, but inconsistently in space and time. A three-year experiment was established in spring 1998 to determine if low rates of picloram or clopyralid applied in combination with S. jugoslavica could enhance S. jugoslavica's capacity to decrease diffuse knapweed populations. The experiment was conducted at three sites in Colorado where S. jugoslavica previously was released in 1993. Picloram and clopyralid were applied separately at 0, 35, 70, and 140 g ai ha-1 during June or September. Density and cover measurements were collected in permanent quadrats three times during each growing season to determine herbicide effects on the growth of diffuse knapweed. Diffuse knapweed plants were harvested outside the permanent quadrats to determine the percentage of diffuse knapweed plants bearing S. jugoslavica larvae. During the spring following herbicide application, all picloram rates applied the previous June and the 35 g rate of clopyralid also applied in June increased the percentage of plants infested by the S. jugoslavica 25% compared to non-treated controls. None of the herbicide treatments increased the percentage of plants infested by S. jugoslavica compared to the control two years following application. Two years following herbicide application, the June-applied 140 g rate of both herbicides decreased diffuse knapweed bolted plant density by more than 70% compared to non-treated controls. Results show that combining low rates of picloram or clopyralid applied in June with S. jugoslavica improves control of diffuse knapweed without harming S. jugoslavica density. Yellow toadflax ecology: Yellow toadflax is a prevalent invasive weed problem in the United States and Canada. It is especially troublesome in the Intermountain west especially in western Colorado. Using GIS technology, yellow toadflax infestations were mapped in the Ripple Creek drainage of the Flat Tops Wilderness in July 1999. Infested site characteristics were recorded and compared to random, non-infested site characteristics. Sixty-five plots that contained yellow toadflax and 70 non-infested sites were mapped. Each plot was circular (19 m radius) with three randomly located transects radiating from the center of each plot. A 1 m2 quadrat (50 by 200 cm) was placed at a random location along each quadrat from which density and cover of yellow toadflax was taken as well as cover of all other plant species. Site characteristics sampled included slope, aspect, elevation, soil type, C:N ratio, organic matter, vegetation type, species composition, percent cover of existing species, and existence of trails or disturbances. Our preliminary results indicate that yellow toadflax occurrence in the Ripple Creek drainage positively correlated with parks, disturbance, aspect, and number of species per plot. Yellow toadflax patches also correlated with yarrow (Achillea lanulosa), orange sneezeweed (Dugaldia hoopesii), one-headed daisy (Erigeron simplex), heart-leaved buttercup (Ranunculus cardiophyllus), Collomia sp., and strawberries (Fragaria sp.). |
| 2001 | Yellow toadflax ecology: Yellow toadflax (Linaria vulgaris) is a prevalent invasive weed problem in the United States and Canada. It is particularly troublesome in the Intermountain west especially in western Colorado. The experimental design was a case paired study at two locations. Using GIS technology, yellow toadflax infested and non-infested plots were mapped in the Ripple Creek drainage and the Marvine Creek drainage of the Flat Tops Wilderness in July 1999 and 2000, respectively. Infested site characteristics were recorded and compared to non-infested site characteristics. In 1999, 65 plots that contained yellow toadflax were compared to 70 non-infested sites and in 2000, 55 infested plots were compared to 55 non-infested plots. Each plot was circular (19 m radius) with three randomly located transects radiating from the center of each plot. A 1 m2 quadrat (50 by 200 cm) was placed at a random location along each transect from which density and cover of yellow toadflax was taken as well as cover of all other plant species. Site characteristics sampled included slope, aspect, elevation, soil texture, C:N ratio, organic matter, vegetation type, species composition, percent cover of existing species, and existence of trails or disturbances. Data from both locations were combined for analysis. The occurrence of yellow toadflax was positively correlated with a higher number of species in circular plots, slope, the presence of trails, and the presence of open parks. Yellow toadflax was typically found in open parks that were flat to slightly sloping. There usually was a trail associated with infestations as well as a greater total number of plant species. Yellow toadflax was also positively correlated in the quadrats with yarrow (Achillea lanulosa), false dandelion (Agoseris glauca), Collomia (Collomia linearis), and rabbitbrush (Chrysothamnus nauseosus) implying that these plants usually were present in yellow toadflax infestations but not always present in non-infested sites. Yellow toadflax was negatively correlated with Engelmann spruce (Picea engemannii) and subalpine fir (Abies lasiocarpa) and was rarely found in timbered sites. Yellow toadflax was not correlated with aspen (Populus tremuloides) or dandelion (Taraxacum officinale) because these species appeared in the majority of sampled sites whether they were infested with yellow toadflax or not. Yellow toadflax occurrence was correlated with % bareground but there was no correlation with % N, C:N, or disturbance. More bareground was detected in yellow toadflax infested sites but elevated N was not important, which allows us to hypothesize that the greater bareground associated with yellow toadflax patches may allow soil temperature to warm sooner and perhaps to a higher temperature than non-infested sites and this may favor yellow toadflax. This may be very important for the survival of yellow toadflax at higher elevations in Colorado. Elevation of the study areas ranged from about 2700 to 3100 m. Experiments must be designed, however, to test this hypothesis. |
| 2002 | Yellow toadflax ecology: Yellow toadflax (Linaria vulgaris) is a prevalent invasive weed problem in the U.S. and Canada, particularly in the Intermountain west. The experimental design was a case paired study at two locations. Yellow toadflax infested and non-infested plots were mapped with GIS technology in the Ripple Creek and the Marvine Creek drainages of the Flat Tops Wilderness in July 1999 and 2000, respectively. Infested site characteristics were compared to non-infested site characteristics. In 1999, 65 infested plots were compared to 70 non-infested plots and in 2000, 55 infested plots were compared to 54 non-infested plots. Each plot was circular (19 m radius) with three transects radiating from the center. A 1 m2 quadrat (50 by 200 cm) was placed at a random location along each transect from which density and cover of yellow toadflax was taken and cover of all other plant species. Site characteristics included slope, aspect, elevation, soil texture, C:N ratio, organic matter, vegetation type (meadows, margins, timber), species composition, percent cover of existing species, and existence of trails or disturbances. Data from both locations were combined for analysis. Logistic regression with stepwise selection was used to generate a model to predict yellow toadflax occurrence on a particular plot based on that site's physical characteristics. The final model included four variables: two vegetation types (meadows, margins), the presence of trails, and total number of plant species per plot. The model correctly classified 89% of the 242 plots sampled. The model indicates that yellow toadflax is most often found in areas that are open, along trails, and have higher species diversity. The model predicts if a particular plot is found in a meadow, along a trail with more than 17 plant species in an 18 m radius area, the probability of yellow toadflax being present is greater than 50 percent. Trails are an apparent dispersal vector for yellow toadflax. Soil moisture may be increased along trails from runoff, which may give yellow toadflax a better opportunity to establish. Also, the open nature of trails and their hard surface with less resistance to wind than vegetated areas may help with dispersal of winged yellow toadflax seeds by wind. The correlation of yellow toadflax with high plant species diversity is consistent with other research in the Intermountain west and may imply that invaded areas have greater resources than native plant communities can use. Most likely yellow toadflax infestations will start along trails or other areas of high moisture availability and from there it will spread into neighboring meadows, but yellow toadflax likely will not invade heavily forested areas because of limited light and moisture. Our results will give land managers in the Flat Tops a tool to identify and intensify monitoring in areas most susceptible to yellow toadflax invasion. This will increase the chance of early detection when the probability of successful control is greatest. Also, the model can be evaluated outside the Flat Tops to determine whether it will accurately predict yellow toadflax elsewhere. |
| 2003 | Yellow toadflax (Linaria vulgaris) is an invasive weed in the U.S. and Canada, particularly in the Intermountain west. The experimental design was a case paired study. Yellow toadflax infested and non-infested plots were mapped with GIS technology in the Ripple Creek and the Marvine Creek drainages of the Flat Tops Wilderness in July 1999 and 2000, respectively. Infested site characteristics were compared to non-infested site characteristics. Each plot was circular (19 m radius) with three transects radiating from the center. A 1 m2 quadrat (50 by 200 cm) was randomly placed along each transect from which density and cover of toadflax was taken and cover of all other plant species. Site characteristics included slope, aspect, elevation, soil texture, C:N ratio, organic matter, vegetation type (meadows, margins, timber), species composition, percent cover of existing species, and existence of trails or disturbances. We hypothesized that at the 1-m2 scale, that characteristics of toadflax plots were different than plots that had none. We used logistic regression to identify characteristics that were important predictors of toadflax occurrence. The resulting logistic regression model identified total species richness, native species cover, and percent plant litter as important predictors of yellow toadflax infestations. The model containing these parameters correctly predicted the presence of toadflax in 82% of the plots sampled. Toadflax infestations were positively associated with species rich areas and negatively associated with native species cover and percent plant litter. Individual variable analysis indicated that native species richness was higher in toadflax plots than non-toadflax plots at the 1-m2 scale, but the degree of difference was dependant on site. Mean native species richness in Ripple Creek and Marvine Creek infested plots was 6.9 and 6.4/m2, respectively, and 4.7 and 5.8/m2 in non-infested plots. Native species cover was negatively associated with toadflax infestations at Marvine Creek, most likely due to higher densities and percent cover, but was insignificant at the less dense Ripple Creek site. Percent plant litter was consistently lower in toadflax plots than non-toadflax plots. When linear regression analysis was used to identify relationships of toadflax density and percent cover with native and additional exotic species richness and percent cover, we found that increasing yellow toadflax density exhibited a curvilinear relationship with native species richness. Native species richness was positively associated with lower toadflax densities; however, once infestations reached a density of 58 shoots/m2, a negative relationship was observed. In addition, native species cover demonstrated a negative linear relationship with yellow toadflax density. Our results suggest that within the Flat Tops Wilderness of Colorado, at the 1-m2 scale, species rich habitats with higher native species assemblages appeared more susceptible to invasion by yellow toadflax. However, once yellow toadflax becomes established in these species rich areas, and reaches a certain density, native species richness begins to decline. |
| 2004 | Yellow toadflax (Linaria vulgaris) is an invasive weed in the Intermountain west. Toadflax infested and non-infested plots were mapped using GIS in the Ripple Creek and Marvine Creek drainages of the Flat Tops Wilderness in July 1999 and 2000, respectively. Site characteristics were compared in infested and non-infested plots. Site characteristics included slope, aspect, elevation, soil texture, C:N, organic matter, vegetation type (meadows, margins, timber), species composition, cover of existing species, and existence of trails or disturbances. Plots were circular (1134 m2) with 3 transects radiating from the center. A 1 m2 quadrat (50 by 200 cm) was randomly placed along each transect from which toadflax density and cover were taken and cover of all other plant species. Logistic regression generated a model to predict the occurrence of toadflax in large plots (1134 m2) based upon site characteristics. Logistic regression also identified important predictors of toadflax occurrence in quadrats (1 m2). For large plots, the final model included 4 variables: 2 vegetation types (meadows, margins), presence of trails, and total number of plant species. The model correctly classified 89% of the 242 plots sampled and indicates that yellow toadflax is most often found in open areas along trails with higher species diversity. If a particular plot is found in a meadow, along a trail with more than 17 plant species, toadflax will be present more than 50 percent of the time. Trails are an apparent dispersal vector for toadflax. Soil moisture may be increased along trails from runoff, which might give toadflax a better opportunity to establish. Also, the open nature of trails with hard surfaces and less resistance to wind than vegetated areas may aid wind dispersal of winged toadflax seeds. The correlation of toadflax with high plant species diversity is consistent with other research in the Intermountain west and implies that invaded areas have greater resources than native plant communities can use. In quadrats, total species richness, native species cover, and plant litter were important predictors of toadflax infestations. This model correctly predicted the presence of toadflax 82% of the time. Toadflax infestations were positively associated with species rich areas and negatively associated with native species cover and plant litter. Plant litter was consistently lower in toadflax plots than non-toadflax plots. Increasing toadflax density exhibited a curvilinear relationship with native species richness. The latter was positively associated with lower toadflax densities; but, once infestations reached a density of 58 shoots/m2, a negative relationship occurred. Native species cover also showed a negative linear relationship with toadflax density. At the 1-m2 scale, species rich habitats with higher native species assemblages appeared more susceptible to invasion by yellow toadflax. However, once toadflax becomes established in species rich areas, and reaches a certain density, native species richness begins to decline. |
| 2005 | Russian knapweed (Acroptilon repens) was used in a modeling approach to better understand the interactions of soil texture, precipitation, and temperature relative to its dominance on rangeland. Locations of 528 patches were superimposed on climate and soils maps to identify 1 km grid cells known to contain the weed. The status of Russian knapweed within a cell was used as the dependent variable in logistic regression to define the environmental envelope for this species. Russian knapweed was most prevalent on clay and clay loam soils and warmer, drier regions of Colorado (18 to 73 cm annual precipitation, 1 to 12 degrees C mean annual temperature). The model was used to predict the probability of occurrence for Russian knapweed across Colorado and our predictions matched the areas of highest known abundance and also indicated areas of highest risk. |
| Impact | |
| 2000 | Our research on combining low rates of herbicides with Sphenoptera jugoslavica demonstrates that picloram and clopyralid, at the rates applied, did not harm the root beetles. Not only did we demonstrate that these herbicides are directly compatible with S. jugoslavica, we also showed that low rates of picloram or clopyralid could enhance the control of diffuse knapweed by the beetle thereby potentially making its performance more consistent in space and time. |
| 2001 | Our research on yellow toadflax ecology shows that yellow toadflax patches at high elevations in Colorado typically occur in open parks that are flat or gently sloping. Yellow toadflax is rarely found in timbered sites where conifers dominate. Because of this, yellow toadflax may be a good candidate for detection by remote sensing. We found a positive association of yellow toadflax with yarrow, false dandelion, Collomia, and rabbitbrush and land managers may be able to use the combination of these species as indicators of sites potentially vulnerable to invasion by yellow toadflax. Land managers also may use the presence of bareground as a predictive tool for sites that may be invaded by yellow toadflax. |
| 2002 | Our research on yellow toadflax ecology shows that yellow toadflax patches at high elevations in Colorado typically occur in open parks that are flat or gently sloping. Yellow toadflax is rarely found in timbered sites where conifers dominate. Most likely, yellow toadflax infestations will first occur along trails where greater moisture availability from runoff occurs and will spread from there into neighboring open meadows. Trails are a major vector of yellow toadflax spread likely because winged seeds are more easily dispersed by wind because of less wind resistance on the hard surface trails compared to vegetated areas. Our model predicts if a particular plot is found in a meadow, along a trail with more than 17 plant species in an 18 m radius area, the probability of yellow toadflax being present is greater than 50 percent. This information can be used to identify areas likely to be invaded by yellow toadflax and improve the probability of early detection when the opportunity for control is greatest. |
| 2003 | Our results support the hypothesis that areas most likely to be invaded by invasive weeds are those sites where species richness and diversity are high. Our data also indicate that yellow toadflax will displace other plant species, particularly native plant species. Yellow toadflax is unpalatable and not consumed by wildlife or livestock and as its populations continue to expand, the carrying capacity of infested rangeland most likely will be decreased. Our results will give land managers in the Flat Tops a tool to identify and intensify monitoring in areas most susceptible to yellow toadflax invasion. This will increase the chance of early detection when the probability of successful control is greatest. Also, the model can be evaluated outside the Flat Tops to determine whether it will accurately predict yellow toadflax elsewhere. |
| 2004 | Our results support the hypothesis that areas most likely to be invaded by invasive weeds are those sites where species richness and diversity are high. Our data also indicate that yellow toadflax will displace other plant species, particularly native plant species. Yellow toadflax is unpalatable and not consumed by wildlife or livestock and as its populations continue to expand, the carrying capacity of infested rangeland most likely will be decreased. Our results will give land managers in the Flat Tops a tool to identify and intensify monitoring in areas most susceptible to yellow toadflax invasion. This will increase the chance of early detection when the probability of successful control is greatest. Also, the model can be evaluated outside the Flat Tops to determine whether it will accurately predict yellow toadflax elsewhere. |
| 2005 | Our results support the hypothesis that invasion by certain weed species can be correlated with environmental factors such as soil texture, precipitation, and temperature. Knowledge of these factors will help land managers understand areas at greatest risk to invasion by Russian knapweed in particular and decrease the land area where monitoring must occur to detect new invasions. |
| Publications | |
| 2000 |
Sutton, J.R. and Beck, K.G. 2000. Possible correlations between site conditions and the occurrence of yellow toadflax in the Flattops Wilderness area of Colorado. Proc. Wes. Soc. Weed Sci. 53:45-46. Wilson, R.G., Beck, K.G. and Westra, P. 2000. Using herbicides in combination with Sphenoptera jugoslavica to improve the root beetle's control of diffuse knapweed. Proc. Wes. Soc. Weed Sci. 53:48. |
| 2001 |
Wilson, Robert G. III. 2001. Biology and management of diffuse knapweed. M.S. Thesis, Colorado State University. |
| 2002 |
Sutton, J. R. and K. G. Beck. 2002. Environmental factors influencing yellow toadflax distribution in the Flat Tops Wilderness area of Colorado. Proc. Wes. Soc. Weed Sci. 55:24. |
| 2003 |
Sutton, Jason R. 2003. Prediction and characterization of yellow toadflax (Linaria vulgaris Mill.) infestations at two scales in the Flat Tops Wilderness of Colorado. M.S. Thesis. |
| 2004 |
Wilson, R., Beck,K.G., and Westra,P. 2004. The Combined effect of herbicides and Sphenoptera jugoslavica on diffuse knapweed population dynamics and S. jugoslavica reproduction success. Weed Sci. 52:418-423. |
| 2005 |
Goslee, S.C., Beck, K.G., and Peters, D.P.C. 2003. Distribution of Russian knapweed in Colorado: Climate and environmental factors. J. Range Manage. 56:206-212. |