Rangeland Weed Research Articles

Enhancing rangeland weed detection through convolutional neural networks and transfer learning

The detection of weed species in rangeland environments is a challenging task due to various factors such as dense, variable species vegetation, ocular occlusion, and a wide variety of plant morphology. Most research in weed detection, however, focuses on croplands. This research addresses the need for accurate rangeland weed detection models by leveraging convolutional neural network (CNN) models enhanced with transfer learning applied to the DeepWeeds data set taken in situ in regional North Eastern Australia. It investigates the effectiveness of transfer learning across seven popular models, utilizing data augmentation and fine-tuning. The performance of these models was evaluated using accuracy metrics and compared against each other. The results demonstrated that transfer learning, coupled with fine tuning, could be a viable solution for generating efficient weed plant detection models with lower demands on computational resources and smaller datasets, despite the challenging conditions of rangeland environments. EfficientNetV2B1 had the highest classification accuracy of 94.2 ​%, and lowest training times. Moreover, high levels of accuracy were also achieved using InceptionV3, VGG16, and Densenet121, albeit with a training time penalty. This research provides insights into the performance of CNN models in challenging rangeland environments, demonstrates the potential of using transfer learning to enhance weed detection models, and underscores the significance of model selection in agricultural applications of CNNs.

Genomic Features of the Host-Specific Fungal Biocontrol Agent Ramularia crupinae Approved for the Management of the Federally Noxious Weed Crupina vulgaris

Ramularia crupinae is a novel host-specific fungal pathogen of the invasive rangeland weed common crupina ( Crupina vulgaris). The foliar and stem blighting pathogen was originally identified in France and has been approved by the United States Department of Agriculture–Animal and Plant Health Inspection Service as the first biological control agent for the management of common crupina in the western United States. In this study, a chromosome-level genome assembly of R. crupinae 00-010 was performed using PacBio sequencing technology. The assembly had a total size of 37.9 Mb distributed across 14 (telomere to telomere) chromosomes, with a GC content of 47.6%. The genome encoded 10,159 protein-coding genes with 890 predicted secreted proteins that comprised both unique and conserved canonical pathogen effectors. Genome comparisons with closely related pathogen species in the family Mycosphaerellaceae demonstrated family-wide orthologous effector groups and Ramularia-specific orthogroups. This new reference sequence is a valuable genomic tool that will assist in characterizing Ramularia–Crupina interactions and increase our understanding of Ramularia pathogenicity on both crops and weeds. [Formula: see text] The author(s) have dedicated the work to the public domain under the Creative Commons CC0 “No Rights Reserved” license by waiving all of his or her rights to the work worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law, 2024.

Zygogramma bicolorata (Coleoptera: Chrysomelidae): an exotic biocontrol agent of the noxious weed, Parthenium hysterophorus (Asteraceae)

Abstract Parthenium hysterophorus Linnaeus (Asteraceae), commonly known as “congress grass,” is among the 10 worst weeds of India and is considered one of the world’s 100 most harmful invasive species. This species has been introduced accidentally into several countries, including India, and is a major agricultural and rangeland weed in more than 45 countries. It affects crop production, suppressing biodiversity, crop yields, the health of farm animals and human beings, and local and regional economies. Biological control has been found to be a cost-effective management strategy to control the weed, with the leaf beetle, Zygogramma bicolorata Pallister (Coleoptera: Chrysomelidae), considered an effective and safe biocontrol agent. Several studies have investigated the beetle’s biology, reproduction, and other life attributes. This review compiles the existing information about Z. bicolorata biology, growth, development, reproduction, food, and host specificity, and the effects of abiotic and biotic stressors on the insect’s life attributes. We anticipate that this review will helpful for improving strategies for this bioagent’s swift mass rearing and for developing classical biological control design for Parthenium weed.

Reproductive success of the invasive weed Linaria dalmatica and seed loss due to predation by Rhinusa neta in northern California, USA

AbstractLinaria dalmatica (L.) Mill. (Dalmatian toadflax, Plantaginaceae) is a serious invasive weed of rangelands and natural areas in the western USA and Canada. The adventive weevil Rhinusa neta (Germar) (Coleoptera: Curculionidae) is a seed predator of L. dalmatica and has recently spread into northern California, USA. Seed loss due to R. neta predation may impact the reproductive success and population abundance of local host plant populations. The aim of this study was to determine the intensity of seed predation and how it may be influenced by the reproductive success of L. dalmatica at six field sites in northern California in 2019 and 2020. Flower production ranged from 40 to 208 flowers per inflorescence, fruit production ranged from 6 to 57 fruit per inflorescence, and seed production ranged from 171 to 3734 seeds per inflorescence. These values are very low (<20%) compared to L. dalmatica plants growing in other regions. Fruit set, an indication of pollination success, ranged from 6.9 to 30.6% in 2019 and from 10.9 to 23.7% in 2020. Average seed set ranged from 36 to 138 seeds per uninfested fruit. The intensity of attack of R. neta ranged from 0 to 70.4%. Fruit attack appeared random at most locations. Average seed set for infested fruit ranged from 0 to 35 seeds per fruit and did not differ significantly between terminal and branch fruit. The population‐level impact of R. neta on seed production was directly related to the intensity of attack at a location. Intensity of attack ranged from <1 to 70% and resulted in corresponding seed losses of 0 to 62.7% among six locations. Predation intensity is expected to increase and stabilize at much higher levels in the future as R. neta continues its immigration into northern California.

Target-site and non-target-site resistance mechanisms confer multiple resistance to glyphosate and 2,4-D in Carduus acanthoides

Carduus acanthoides L. is mainly a range-land weed, but in the 2010s has begun to invade GM crop production systems in Córdoba (Argentina), where glyphosate and 2,4-D have been commonly applied. In 2020, C. acanthoides was found with multiple resistance to these two herbicides. In this study, the mechanisms that confer multiple resistance to glyphosate and 2,4-D, were characterized in one resistant (R) population of C. acanthoides in comparison to a susceptible (S) population. No differences in 14C-herbicide absorption and translocation were observed between R and S populations. In addition, 14C-glyphosate was well translocated to the shoots (∼30%) and roots (∼16%) in both R and S plants, while most of 14C-2,4-D remained restricted in the treated leaf. Glyphosate metabolism did not contribute to resistance of the R population; however, as corroborated by malathion pretreatment, the mechanism of resistance to 2,4-D was enhanced metabolism (63% of the herbicide) mediated by cytochrome P450 (Cyt-P450). No differences were found in baseline EPSPS activity, copy number, and/or gene expression between the R and S populations, but a Pro-106-Ser mutation in EPSPS was present in the R population. Multiple resistances in the R population of C. acanthoides from Argentina were governed by target site resistance (a Pro-106 mutation for glyphosate) and non-target site resistance (Cyt-P450-based metabolic resistance for 2,4-D) mechanisms. This is the first case of resistance to glyphosate and 2,4-D confirmed for this weed in the world.

Weed Management in Pastures and Rangeland—2022

This publication provides practical pasture and rangeland weed management information. Written by Brent Sellers and Pratap Devkota, and published by the UF/IFAS Agronomy Department, revised January 2022.

Robotic Spot Spraying of Harrisia Cactus (Harrisia martinii) in Grazing Pastures of the Australian Rangelands.

Harrisia cactus, Harrisia martinii, is a serious weed affecting hundreds of thousands of hectares of native pasture in the Australian rangelands. Despite the landmark success of past biological control agents for the invasive weed and significant investment in its eradication by the Queensland Government (roughly $156M since 1960), it still takes hold in the cooler rangeland environments of northern New South Wales and southern Queensland. In the past decade, landholders with large infestations in these locations have spent approximately $20,000 to $30,000 per annum on herbicide control measures to reduce the impact of the weed on their grazing operations. Current chemical control requires manual hand spot spraying with high quantities of herbicide for foliar application. These methods are labour intensive and costly, and in some cases inhibit landholders from performing control at all. Robotic spot spraying offers a potential solution to these issues, but existing solutions are not suitable for the rangeland environment. This work presents the methods and results of an in situ field trial of a novel robotic spot spraying solution, AutoWeed, for treating harrisia cactus that (1) more than halves the operation time, (2) can reduce herbicide usage by up to 54% and (3) can reduce the cost of herbicide by up to $18.15 per ha compared to the existing hand spraying approach. The AutoWeed spot spraying system used the MobileNetV2 deep learning architecture to perform real time spot spraying of harrisia cactus with 97.2% average recall accuracy and weed knockdown efficacy of up to 96%. Experimental trials showed that the AutoWeed spot sprayer achieved the same level of knockdown of harrisia cactus as traditional hand spraying in low, medium and high density infestations. This work represents a significant step forward for spot spraying of weeds in the Australian rangelands that will reduce labour and herbicide costs for landholders as the technology sees more uptake in the future.

Post-release genetic assessment of two congeneric weed biological control agents

Introductions of biological control agents are unique field experiments to examine the response of novel organisms to new environments. Post-release monitoring is particularly challenging where closely related and morphologically similar biological control agents have been released. Two closely-related phytophagous weevils, Larinus minutus and L. obtusus, were introduced as biological control agents for two major rangeland weeds, diffuse knapweed, Centaurea diffusa and spotted knapweed, C. stoebe subsp. micranthos in North America. The release of the Larinus species coincided with a decline in C. diffusa abundance in many areas. However, it was not clear whether both species of Larinus had become established as they are morphologically very similar. We asked: (1) Could genetic markers be used to identify both the weevil species? (2) Do both weevil species attack both knapweed species in Canada? (3) Are the levels of genetic diversity of the introduced populations of the weevils similar to those of a European source population? (4) Is genetic mixing between the weevil species occurring? Both microsatellite and CO1 mtDNA markers distinguish between the two weevil species. Larinus obtusus was only found on spotted knapweed, while L. minutus was more widely distributed and attacked both weed species. The relatively large initial introductions of weevils to British Columbia (4,300 L. minutus and 5,500 L. obtusus) together with rapid population growth and frequent redistribution among sites has been sufficient to maintain levels of genetic diversity. Results from small samples showed that haplotype, nucleotide diversity and allelic richness of the introduced populations were comparable to those of a European population. We found no evidence for hybridization in the introduced populations.

Seed germination response of a potential rangeland weed Psilocaulon granulicaule to selected environmental conditions

Studies show that just over 620 non-native naturalised plant species have been recorded within the Australian rangelands, some of which have a capacity to cause significant impacts on rangeland flora and grazing activity. Although Psilocaulon granulicaule (Haw.), Schwantes is listed as a highly invasive environmental weed species, there has been no previous research into its seed ecology. Therefore, this study was conducted to investigate the effects of temperature, light, pH, water stress, heat-shock, and salinity on the germination of P. granulicaule. In this study, four temperature regimes covering four different day and night temperature variations (17–7°C, 25–15°C, 30–25°C and 40–30°C) and two light regimes (12-h light–12-h dark, 24-h dark) were investigated. The effects of pH, water stress, heat-shock and salinity were investigated, using pH buffers, polyethylene glycol solutions, three heat shock events under four temperatures and a range of NaCl solutions. These tests were conducted under the identified optimum temperature range (25/15°C) and light regime for seed germination. The results showed that both temperature and photoperiod significantly influenced the germination rate, with 94.2% germination in the 25–15°C range under a 12-h light-12-h dark regime. Higher temperatures (30–40°C) reduced seed germination to <58% germination in both light regimes (57.5%, 12-h light-12-h dark; 54.17%, 24-h dark). The highest germination rates were observed in low pH solutions, high moisture levels, low heat-shocks and low salinity. The study showed that this species is sensitive to environmental factors such as temperature, light, pH, moisture, heat shock and salinity, suggesting that these factors can be used as critical indicators to guide effective management practices to address this weed problem. Given that seeds are sensitive to radiant heat, burning could be used as a tool to effectively manage this species.

Land Manager Perceptions of Opportunities and Constraints of Using Livestock to Manage Invasive Plants

The ecological impacts of rangeland invasive plants have been widely documented, but the social aspects of how managers perceive their impacts and options for control have been relatively understudied, and successful, long-term invasive plant management programs are limited. In particular, though a growing body of research has identified livestock grazing as the most practical and economical tool for controlling invasive rangeland plants, to date there has not been a systematic assessment of the challenges and opportunities producers and other land managers see as most important when considering using livestock to manage invasive plants. In-depth, semistructured interviews with California annual grass and hardwood rangeland ranchers, public agency personnel, and nongovernmental organization land managers were used to address this need. Although interviewees broadly agreed that grazing could be an effective management tool, differences emerged among the three groups in how they prioritized invasive plant control, the amount of resources devoted to control, and the grazing strategies employed. Interviewees identified key challenges that hinder broad-scale adoption of control efforts, including the potential incompatibility of invasive plant management and livestock production; a lack of secure, long-term access to land for many ranchers; incomplete or insufficient information, such as the location or extent of infestations or the economic impacts to operations of invasive plants; and the temporal and spatial variability of the ecosystem. By identifying key socioecological drivers that influence the degree to which livestock are used to manage invasive plants, this study was able to identify potential pathways to move our growing understanding of the science of targeted grazing into practice. Research, extension, and grazing programs that address these barriers should help increase the extent to which we can effectively use livestock to slow and perhaps reverse the spread of some of our most serious rangeland weeds.

Performance of the biocontrol agent Secusio extensa (Lepidoptera: Erebidae) on its target host, Senecio madagascariensis (Madagascar fireweed), on an alternate host, Delairea odorata (Cape ivy), and on non-target plants, in Hawaii

The lepidopteran herbivore Secusio extensa (Erebidae: Arctiinae) was purposefully released in Hawaii in 2013 to control Madagascar fireweed, Senecio madagascariensis (Asteraceae: Senecioneae), an invasive weed of rangelands and other open areas. In conjunction with this biocontrol effort, we conducted pre- and post-release monitoring of S. madagascariensis in field plots, and post-release monitoring of non-target plants. Secusio extensa successfully established on the two islands where it was released, dispersed widely, and even colonized a neighboring island. However, in the first three years after release, S. extensa could only be found sporadically and in low abundances on its target host. In contrast, it underwent population outbreaks on the related, alternative host, Cape ivy (Delairea odorata, Asteraceae: Senecioneae), which is also a noxious weed in Hawaii. In addition to D. odorata, we observed S. extensa to feed on only two other related non-target plants in the tribe Senecioneae, among 24 species monitored. In several microsites supporting higher densities of S. extensa on S. madagascariensis, larval abundances peaked during a short period in the spring, but were insufficiently high to cause substantial damage to affected plants. The causes for low larval densities of S. extensa on its target host are unclear. We found no strong evidence for parasitism or predation of S. extensa larvae. However, females preferred to oviposit on D. odorata over S. madagascariensis, and larvae exhibited a marginally significant preference for feeding on D. odorata over S. madagascariensis. This preference, along with lower nutritional quality of S. madagascariensis relative to D. odorata, may explain why S. extensa has been unable to build large populations on its intended target, but has done so on an alternate host. Clarifying the roles of these mechanisms may help predict the continuing performance of S. extensa in controlling S. madagascariensis.

Tolerance of the eriophyid mite Aceria salsolae to UV-A light and implications for biological control of Russian thistle.

Aceria salsolae (Acari: Eriophyidae) is being evaluated as a candidate biological control agent of Russian thistle (Salsola tragus, Chenopodiaceae), a major invasive weed of rangelands and dryland crops in the western USA. Prior laboratory host range testing under artificial lighting indicated reproduction on non-native Bassia hyssopifolia and on a native plant, Suaeda calceoliformis. However, in field tests in the native range, mite populations released on these 'nontarget' plants remained low. We hypothesized that UV-A light, which can affect behavior of tetranychid mites, would affect populations of the eriophyid A. salsolae differently on the target and nontarget plant species, decreasing the mite's realized host range. Plants were infested with A. salsolae under lamps that emitted UV-A, along with broad-spectrum lighting, and the size of mite populations and plant growth was compared to infested plants exposed only to broad-spectrum light. Russian thistle supported 3- to 55-fold larger mite populations than nontarget plants regardless of UV-A treatment. UV-A exposure did not affect mite populations on Russian thistle or S. calceoliformis, whereas it increased populations 7-fold on B. hyssopifolia. Main stems on nontarget plants grew 2- to 6-fold faster than did Russian thistle under either light treatment. The two nontarget plants attained greater volume under the control light regime than UV-A, but Russian thistle was unaffected. Although Russian thistle was always the superior host, addition of UV-A light to the artificial lighting regime did not reduce the ability of A. salsolae to reproduce on the two nontarget species, suggesting that UV-B or other environmental factors may be more important in limiting mite populations in the field.

Host Associations of Gall-InducingProdiplosis longifila(Diptera: Cecidomyiidae) from Bolivia: Implications for Its Use as a Biological Control Agent forJatropha gossypiifolia(Euphorbiaceae)

Based on field host range and damage potential, we explored the prospects of exploiting a gall midge from Jatropha clavuligera Mull. Arg. (Euphorbiaceae) in Bolivia as a “new-association” biological control agent for Jatropha gossypiifolia L. (Euphorbiaceae), a major rangeland weed in Australia. The gall midge, determined morphologically as Prodiplosis longifila Gagne (Diptera: Cecidomyiidae), induces rosette galls, resulting in shoot tip dieback in J. clavuligera in Bolivia. Although P. longifila is reported to occur on a range of crops in the Neotropics, its feeding on J. clavuligera in Bolivia is strikingly different in that it induces rosette galls on its shoots. In other countries, P. longifila larvae feed on leaves, buds, flowers and fruits, with no gall symptoms. There was no evidence of P. longifila-induced galls on crop plants (reported hosts in other countries), other species of Jatropha, or other species of Euphorbiaceae in Bolivia, and in choice tests conducted in the quarantine facility in South Africa. Bolivian entomological records do not report any damage to crops by P. longifila. We propose that P. longifila in Bolivia is possibly a new species restricted to a few closely related species of Jatropha and a part of a cryptic species complex. Based on the susceptibility of J. gossypiifolia to the gall-inducing P. longifila in the field in Bolivia and in quarantine tests in South Africa, we propose that the gall-inducing P. longifila sourced from J. clavuligera in Bolivia is a potential “new-association” biological control agent for J. gossypiifolia in Australia.

A review of the integration of classical biological control with other techniques to manage invasive weeds in natural areas and rangelands

Integrating classical biological control with other management techniques such as herbicide, fire, mechanical control, grazing, or plant competition, can be the most effective way to manage invasive weeds in natural areas and rangelands. Biological control agents can be protected from potential negative impacts of these weed control methods through untreated refugia or by applying the treatment at a time when the agent is not vulnerable. A literature review of experiments that integrated biological control with other management strategies from 1987 to 2017 yielded 39 terrestrial and 16 aquatic studies. The tactics most frequently integrated with biological control were herbicide applications and plant competition. Despite numerous examples of successful programs and calls for more widespread integration of biological control with other weed management strategies, there was no increase in the number of studies reported annually over time. Additional studies investigating the ecological and economic benefits of integrated weed management are needed.

Pre-emergence Control of Six Invasive Winter Annual Grasses with Imazapic and Indaziflam

Managing invasive winter annual grasses on noncrop and rangeland remains a constant challenge throughout many regions of the United States. Currently, there are limited management options for controlling winter annual grasses that work consistently, provide multiple years of control, and do not injure desirable plant communities. Imazapic has been one of the most widely used herbicides for downy brome control on rangeland; however, control with imazapic has been inconsistent beyond the application year and perennial grass injury is not uncommon. Indaziflam, a new herbicide mode of action for rangeland weed management, has shown promise in providing long-term downy brome control. A greenhouse study was conducted to compare pre-emergence activity of imazapic and indaziflam on six invasive winter annual grasses: downy brome, cereal or feral rye, jointed goatgrass, Japanese brome, medusahead, and ventenata. For both herbicides, seven rates were used to develop dose-response curves for each species. Log-logist...

Potential disease agents in domestic goats and relevance to bighorn sheep (Ovis canadensis) management.

Domestic goats are raised for meat, milk and hair production, in herds for rangeland weed control, and as pack animals. Domestic sheep, goats and wild bighorn sheep are all susceptible to a multifactorial pneumonia. We sampled 43 herd goats from 7 herds and 48 pack goats from 11 herds for viral and bacterial serology, parasitology, and Pasteurellaceae microbiology. The goats in this study were in generally good health, although most goats did harbor various pathogens and parasites including several bacteria, specifically Pasteurellaceae, which have been associated with pneumonia in free-ranging bighorn sheep. It is not known if domestic goats can transmit the Pasteurellaceae or other pathogens found in this study readily to wild bighorn sheep. However, due the possibility of transmission, domestic goats in areas in or near bighorn sheep habitat should be managed to minimize the risk of spreading disease agents to bighorn sheep.

Sierra Foothill REC: Quantifying IPM benefits in rangeland systems

News from the RECs Sierra Foothill REC: Quantifying IPM benefits in rangeland systems T here’s a strong scientific case that integrated pest management strategies for rangeland weeds are effective — yet adoption of these practices on rangeland is extremely low. How come? One likely reason, says Jeremy James, director of Sierra Foothill Research and Extension Center (SFREC), is that it’s unclear whether investing in reducing weed populations makes financial sense. Microsoft, ESRI 174 CALIFORNIA AGRICULTURE • VOLUME 70 , NUMBER 4 Elena Zhukova running a series of experi- ments at SFREC. Beginning in 2013, the team manipulated pastures to have different degrees of weediness, measured by the percentage of land infested with medusahead, a com- mon rangeland weed. After 3 years, they had developed Medusahead (in green, above) is an a series of 13 experimental aggressive invasive annual grass that is 5-acre paddocks, with me- unpalatable to livestock for most of its “We have a good idea that invasive plants have negative dusahead coverage ranging life cycle. ecological impacts — but what we surprisingly do not know is from 11% to 50%. if there is actually any economic cost of having weeds on range- This spring, the researchers stocked the pastures with steers lands,” James said. from March through May and evaluated how weight gain var- A national assessment of rangeland integrated pest manage- ied with weed coverage. ment strategies like Results from this first season of tests indicate that weight targeted grazing, seed- gain does indeed suffer as weediness increases: for every N ing of desirable plants, 10% increase in weed cover, total weight gain was reduced by prescribed fire and the roughly 30 pounds per acre — a reduction in market value of use of selective herbi- $30 to $50 per acre at current prices. cides found abundant “For beef cattle production, those numbers are right in the evidence that these prac- middle of being something that might pencil out to treat,” said tices work (Sheley et al. James. “They’re not going to cause ranchers to go out of busi- 2011). However, the same ness, but at the same time they’re not negligible.” study also found that The study is scheduled to continue for at least another 2 that these practices are years. The team will replicate the spring grazing study and also little-used — in part be- assess the effects of weeds at other times of year, James said. cause of the difficulty of The project is part of a collection of studies funded by the assessing their economic U.S. Department of Agriculture that aims to quantify the value benefit. of the many ecosystem services that rangelands can provide. To better understand Some of these, like the effect of weeds on cattle weight gain, can the economic case for be directly quantified in economic terms. Others — such as the weed management, provision of habitat to support biodiversity or the benefits that James and a team of healthy soil can provide for groundwater infiltration — may be researchers — including more difficult to value but may still be desirable for ranchers to Tehama County live- support. The overall goal of the project, James said, is to docu- stock, range and natural ment the full suite of benefits that investments in rangeland resources advisor Josh management can deliver. c Paddock borders Davy, Stanislaus County — Jim Downing range and natural re- sources advisor Theresa Becchetti, Shasta County Reference livestock, range and Meters Sheley RL, James JJ, Rinella MJ, et al. 2011. Invasive plant management on antici- natural resources advi- pated conservation benefits: A scientific assessment. Chapter 7. In: Briske DD (ed.). The 13 test paddocks (nine of which are Conservation Benefits of Rangeland Practices: Assessment, Recommendations, and sor Larry Forero and UC shown here) enclose 5 acres each and Knowledge Gaps. US Department of Agriculture, Natural Resources Conservation Davis Plant Science pro- are infested with medusahead to varying Service. bit.ly/2c1H0s2. degrees — from 11% to 50% coverage. fessor Emilio Laca — are

Seed fall, seed predation, twigging and litter fall of Cascabela thevetia (L.) Lippold

Cascabela thevetia (L.) Lippold is a garden ornamental in northern Australia and two biotypes, the peach and the yellow, are recognised. In some areas it has naturalised and now has environmental and economic effects. As part of a broader research program into the ecology of C. thevetia, a field study was undertaken in northern Queensland to quantify seed fall and seed predation (by avian wildlife) of the peach biotype. The amount of twigging caused by birds while they fed on the seeds and the level of litter production were also recorded. Seed fall, seed predation, twigging and litter production occurred in all months of the year. Seed fall increased slowly over late spring and summer before peaking in mid- to late autumn (April–May) and then declining until October. Mean (± s.e.m.) estimated total annual seed fall was 19140 ± 2880 and 17030 ± 2930 seeds ha–1 in the first and second years respectively. Seed predation by birds was substantial, with 57% of all seeds predated. Birds also chewed an average of 600 twigs ha–1 year–1. Litter production varied from 430 to 950 kg dry weight (DW) ha–1 month–1. In total, 7900 ± 640 and 7390 ± 1420 kg (DW) litter was produced during the first and second years respectively. Although seed production of C. thevetia is less than a lot of other rangeland weeds, seed predation by birds further reduces the number of seeds entering the soil seed bank. The stem damage that occurred in conjunction with seed predation contributed to overall litter production and warrants further investigation in terms of its effect on plant growth.

Nontarget herbivory by a weed biocontrol insect is limited to spillover, reducing the chance of population-level impacts.

Insects approved for classical biocontrol of weeds are often capable of using close relatives of their target weed for feeding, oviposition, or larval development, with reduced preference and performance. When nontarget herbivory occurs and is suspected to reduce survival, growth, or fecundity of individual plants, and insects are capable of reproducing on their nontarget host, characterization of spatial and temporal patterns of the occurrence and intensity of herbivory is valuable for predicting potential population-level effects. Here, we perform a novel post-release manipulative field experiment with a root-feeding biocontrol weevil, Mogulones crucifer, released in Canada to control the rangeland weed Cynoglossum officinale, to test for its ability to establish on the nontarget plant Hackelia micrantha. After Cynoglossum, M. crucifer exhibits its highest preference for and performance on Hackelia spp. We released M. crucifer on Canadian rangeland sites with naturally occurring populations of H. micrantha growing interspersed with the target weed or in the near absence of the target weed. Adult weevil feeding on surrounding plants was monitored for three summers after release (years 0, 1, and 2), and, subsequently, subsets of plants were destructively sampled to determine M. crucifer oviposition levels. Additional oviposition and larval development data were obtained from seven non-experimental sites where weevils were released zero, three, or four years earlier. M. crucifer was not detected on experimental sites without C. officinale after two years, and nontarget herbivory was restricted to rare, low-level spillover. Visible evidence of adult herbivory (i.e., scars on shoots) was associated with oviposition in 90% of targets but only 30% of nontarget plants. We infer, through ecological refuge theory, that nontarget population-level impacts from M. crucifer spillover are unlikely because of temporal, spatial, and probabilistic refuges from herbivory, and make recommendations for monitoring and management of biocontrol systems with similar attributes, such as removing target plants around nontarget populations of interest. Because M. crucifer is among the least host-specific of the modern weed biocontrol agents, and H. micrantha is likely one of its most highly preferred nontargets, these conclusions are, arguably, generally applicable to other nontarget plants and biocontrol systems.

Identification of the Infection Route of a Fusarium Seed Pathogen into Nondormant Bromus tectorum Seeds.

The genus Fusarium has a wide host range and causes many different forms of plant disease. These include seed rot and seedling blight diseases of cultivated plants. The diseases caused by Fusarium on wild plants are less well-known. In this study, we examined disease development caused by Fusarium sp. n on nondormant seeds of the important rangeland weed Bromus tectorum as part of broader studies of the phenomenon of stand failure or "die-off" in this annual grass. We previously isolated an undescribed species in the F. tricinctum species complex from die-off soils and showed that it is pathogenic on seeds. It can cause high mortality of nondormant B. tectorum seeds, especially under conditions of water stress, but rarely attacks dormant seeds. In this study, we used scanning electron microscopy (SEM) to investigate the mode of attack used by this pathogen. Nondormant B. tectorum seeds (i.e., florets containing caryopses) were inoculated with isolate Skull C1 macroconidia. Seeds were then exposed to water stress conditions (-1.5 MPa) for 7 days and then transferred to free water. Time lapse SEM photographs of healthy versus infected seeds revealed that hyphae under water stress conditions grew toward and culminated their attack at the abscission layer of the floret attachment scar. A prominent infection cushion, apparent macroscopically as a white tuft of mycelium at the radicle end of the seed, developed within 48 h after inoculation. Seeds that lacked an infection cushion completed germination upon transfer to free water, whereas seeds with an infection cushion were almost always killed. In addition, hyphae on seeds that did not initiate germination lacked directional growth and did not develop the infection cushion. This strongly suggests that the fungal attack is triggered by seed exudates released through the floret attachment scar at the initiation of germination. Images of cross sections of infected seeds showed that the fungal hyphae first penetrated the caryposis wall, then entered the embryo, and later ramified throughout the endosperm, completely destroying the seed.