Thomas. L. Marsh

The economic value of cultivars resistant to disease is of great interest, but how growers change their fungicide use in response to host resistance may be nuanced. We draw upon a well-described data set of the incidence of hop plants with powdery mildew and associated production metadata and demonstrate the utility of Bayesian networks as a framework for quantifying causal relationships for fungicide use and cost in response to host resistance. Conditional Gaussian Bayesian network models applied to cultivars differing in race-specific resistance to powdery mildew revealed cultivar resistance to powdery mildew influenced disease levels in early spring, which had a causal effect on how often and what fungicides growers later applied. Annual costs depended on not only the number of applications made but also the specific types of fungicides growers selected. Fungicide costs were little changed on cultivars that possessed race-specific resistance to only one of two extant strains of the pathogen. For cultivars with resistance to both pathogen strains, annual costs of fungicides were reduced commensurate with the level of resistance. Predicted values from the Bayesian networks and simulation indicate that growers apply a baseline level of fungicide, independent of cultivar resistance. Fungicide cost savings result from how fungicide inputs differentially scale with the incidence of powdery mildew and the type of fungicides used. Our analyses indicate that for a high-value crop, deployment of disease resistance may cause complex and unexpected changes in growers’ fungicide use patterns that may not be obvious in simplified randomized controlled trials.

Disease management may require multiple decisions by growers on whether to apply a pesticide, the frequency of applications, and also the type of pesticide to use. We developed models for estimating the probability of switching behavior by hop growers related to use of nonsynthetic or synthetic fungicides, or mixtures thereof, for hop powdery mildew (Podosphaera macularis). Growers used nonsynthetic fungicides alone in 61.5% of their applications, later switching to synthetic fungicides or mixtures of the two in 21.5 or 17.0% of their total applications, respectively. Binary logistic regression predicted that the likelihood of switching fungicide type was associated with use of a nonsynthetic fungicide in the first application of the year and how early in the season growers made that application. The predicted probability of switching increased proportionate to the incidence of plants with powdery mildew but with a weaker effect. Multinomial logistic regression modeling of switching between nonsynthetic, synthetic, or mixtures of fungicides indicated a complex pattern of switching behavior. A random forest classification algorithm identified the most important variables in the multinomial logistic regression model as time of year, individual grower effects, the date of the first application, seasonal mean disease incidence, thoroughness of spring pruning, and cultivar susceptibility to two races of the fungus. Overall, our analyses indicate that growers switch from nonsynthetic fungicides to more active (and more expensive) synthetic fungicides or mixtures thereof at critical periods of crop susceptibility and when expected or observed disease incidence increases. Reducing the use of synthetic fungicide inputs may be closely linked to disease risk mitigation, as risk mitigation is indissolubly linked to growers’ decisions to switch to these more potent fungicides or fungicide mixtures.

Plant disease monitoring metadata are a rich source of information for predicting pesticide use and costs on individual fields, farms, or regionally. Pesticide use patterns for management of hop powdery mildew (Podosphaera macularis) in Oregon were summarized by frequency of use of active ingredients and Fungicide Resistance Action Committee (FRAC) code. There was extensive variation in the frequency of use of specific active ingredients and FRAC codes among growers, ranging from a mean of 0 to 2.41 applications per yard. All growers used fungicides with FRAC codes 7/11 and 13; all but one grower used fungicide FRAC codes M02, NC, 3, and 11. Among fungicides with a moderate to high risk of developing resistance (FRAC codes 3, 5, 7, 7/11 premix, 11, or 13), the mean number of applications was ≤1.35 per yard across all growers and never exceeded four applications within a given yard. Hierarchical cluster analysis identified that the diverse use patterns could be categorized into three groups, which we define as pesticide programs. We fitted a random forest regression model and used Shapley values to quantify the importance of fungicide program in predicting the number of active ingredients applied and their costs relative to other known factors that influence these outcomes in individual yards. Pesticide program was the third most important variable predicting annual costs and the fifth most important predictor for the number of active ingredients applied. We also found hints that growers may switch between one of the three pesticide programs depending on the incidence of powdery mildew.

Methods for causal inference from observational data are common in human disease epidemiology and social sciences but are used relatively little in plant pathology. We draw upon an extensive data set of the incidence of hop plants with powdery mildew (caused by Podosphaera macularis) collected from yards in Oregon from 2014 to 2017 and associated metadata on grower cultural practices, cultivar susceptibility to powdery mildew, and pesticide application records to understand variation in and causes of growers’ fungicide use and associated costs. An instrumental causal forest model identified growers’ spring pruning thoroughness, cultivar susceptibility to two of the dominant pathogenic races of P. macularis, network centrality of yards during May–June and June–July time transitions, and the initial strain of the fungus detected as important variables determining the number of pesticide active constituents applied by growers and the associated costs they incurred in response to powdery mildew. Exposure-response function models fit after covariate weighting indicated that both the number of pesticide active constituents applied and their associated costs scaled linearly with the seasonal mean incidence of plants with powdery mildew. Although the causes of pesticide use intensity are multifaceted, biological and production factors collectively influence the incidence of powdery mildew, which has a direct exposure-response relationship with the number of pesticide active constituents that growers apply and their costs. Our analyses point to several potential strategies for reducing pesticide use and costs for management of powdery mildew on hop. We also highlight the utility of these methods for causal inference in observational studies.

Vaccination for foot-and-mouth (FMD) disease remains low in parts of Africa despite the existence of vaccines. In East Africa, the presence of multiple virus serotypes and sub-types makes matching a vaccine with the circulating virus type in the field, or providing a high potency vaccine, a challenge. In this paper we use game theory to show that the resulting vaccine uncertainty associated with these vaccination conditions in an endemic setting help explain the low vaccine uptake. We evaluate vaccination for FMD in the context of East Africa due to FMD being endemic in the region, the diversity of FMD virus types, and barriers to implementing other disease control measures, such as controlled movements. We incorporate these conditions into a vaccination game setting and compare the payoffs to those of a traditional vaccination game for seasonal influenza and commercial livestock vaccination in a developed country context. In showing that vaccination provides households with a lower payoff than not vaccinating, our simple game theoretical explanation supports existing evidence calling for improved vaccine quality and efforts to enhance surveillance to provide early information on disease status.

The market structure and recipes for beer has been rapidly changing with craft beers attracting more consumers. Perceived hops quality (hoppiness) is one of the main attributes that microbrewers alter to differentiate their products to satisfy consumers’ changing tastes and preferences. We hypothesize that, in addition to manipulating beer-processing conditions, the conditions under which the hops are grown may also influence the final sensory properties of the beer. Using hops from a field experiment coupled with sensory attributes and sociodemographic characteristics from a contingent valuation survey, we analyzed the impact of under-fertilized hop treatments during the growing season on consumers’ willingness to pay for beer. The results indicate that uninformed consumers in a blind tasting could identify the differences in beer made from hops across the fertilization treatments and, thus, implying that all else equal sufficient fertilizer is required to achieve satisfactory hoppiness for which consumers are willing to pay. (JEL Classifications: C91, D12, L66, Q11)

African pastoralists’ decision to vaccinate cattle generates significant household income, translating into broad societal goals.

A foot‐and‐mouth disease outbreak, although having a low probability of occurrence, results in losses of export markets and introduces considerable potential disease management and eradication costs. We develop a dynamic model that integrates beef cattle production, disease dissemination, domestic consumption, and international trade and captures the intertemporal economic welfare impacts of mitigation measures. The model is applied to the case of a hypothetical outbreak in Canada to capture changes in producer profits, consumer price, and government costs due to the outbreak and sums these to measure changes in total economic welfare of the beef industry. Mitigation scenarios are reported for stamping‐out, movement controls, vaccination, and preemptive slaughter.