The enemy release hypothesis proposes that invasion by exotic plant species is driven by their release from natural enemies (i.e. herbivores and pathogens) in their introduced ranges. However, in many cases, natural enemies, which may be introduced or managed to regulate invasive species, may fail to impact target host populations. Landscape heterogeneity, which can affect both the population dynamics of the pathogen and the susceptibility and the density of hosts, may contribute to why pathogens fail to control hosts despite established negative disease impacts. We explored patterns of post‐fire infection of the fungal head‐smut pathogen Ustilago bullata on the invasive annual cheatgrass Bromus tectorum, which has caused the notorious grass‐fire cycle and ecosystem degradation across Western North America. We asked whether infection level was a driver of host density or vice‐versa, and how weather affected infection and how spatial patterns of infection varied with time since fire, using a combination of structural equation modelling (SEM), proportional odds modelling and entropy‐based local indicator of spatial association (ELSA) on data from >700 plots spanning >100,000 ha remeasured annually for 4 years. Observed infection levels increased with greater prior‐year cheatgrass cover, and disease severity did not suppress cheatgrass populations. Warm, humid fall/winters and proximity to fire refugia (unburned patches) were associated with more infections. Infection clustering was most evident 2–3 years following fire with warm‐wet fall–winter conditions and decreased after two drier, colder winters. Synthesis. Severity of fungal disease did not result in measurable reductions of populations of a non‐native, invasive host species, cheatgrass, which suggests that natural enemies may not strongly regulate cheatgrass in its introduced range. Landscape heterogeneity associated with disturbance and weather limited population‐level infection of hosts by the fungal pathogen. Disturbance (specifically wildfire) and variable weather are key components of this and similar invasion systems, and likely need to be considered when evaluating disease dynamics and potential for natural enemies to influence invasion potential.

Altered climate, including weather extremes, can cause major shifts in vegetative recovery after disturbances. Predictive models that can identify the separate and combined temporal effects of disturbance and weather on plant communities and that are transferable among sites are needed to guide vulnerability assessments and management interventions. We asked how functional group abundance responded to time since fire and antecedent weather, if long-term vegetation trajectories were better explained by initial post-fire weather conditions or by general five-year antecedent weather, and if weather effects helped predict post-fire vegetation abundances at a new site. We parameterized models using a 30-yr vegetation monitoring dataset from burned and unburned areas of the Orchard Training Area (OCTC) of southern Idaho, USA, and monthly PRISM data, and assessed model transferability on an independent dataset from the well-sampled Soda wildfire area along the Idaho/Oregon border. Sagebrush density increased with lower mean air temperature of the coldest month and slightly increased with higher mean air temperature of the hottest month, and with higher maximum January–June precipitation. Perennial grass cover increased in relation to higher precipitation, measured annually in the first four years after fire and/or in September–November the year of fire. Annual grass increased in relation to higher March–May precipitation in the year after fire, but not with September–November precipitation in the year of fire. Initial post-fire weather conditions explained 1% more variation in sagebrush density than recent antecedent 5-yr weather did but did not explain additional variation in perennial or annual grass cover. Inclusion of weather variables increased transferability of models for predicting perennial and annual grass cover from the OCTC to the Soda wildfire regardless of the time period in which weather was considered. In contrast, inclusion of weather variables did not affect transferability of the forecasts of post-fire sagebrush density from the OCTC to the Soda site. Although model transferability may be improved by including weather covariates when predicting post-fire vegetation recovery, predictions may be surprisingly unaffected by the temporal windows in which coarse-scale gridded weather data are considered.