The role of citizen science in addressing grand challenges in food and agriculture research.

The power of citizen science to contribute to both science and society is gaining increased recognition, particularly in physics and biology. Although there is a long history of public engagement in agriculture and food science, the term 'citizen science' has rarely been applied to these efforts. Similarly, in the emerging field of citizen science, most new citizen science projects do not focus on food or agriculture. Here, we convened thought leaders from a broad range of fields related to citizen science, agriculture, and food science to highlight key opportunities for bridging these overlapping yet disconnected communities/fields and identify ways to leverage their respective strengths. Specifically, we show that (i) citizen science projects are addressing many grand challenges facing our food systems, as outlined by the United States National Institute of Food and Agriculture, as well as broader Sustainable Development Goals set by the United Nations Development Programme, (ii) there exist emerging opportunities and unique challenges for citizen science in agriculture/food research, and (iii) the greatest opportunities for the development of citizen science projects in agriculture and food science will be gained by using the existing infrastructure and tools of Extension programmes and through the engagement of urban communities. Further, we argue there is no better time to foster greater collaboration between these fields given the trend of shrinking Extension programmes, the increasing need to apply innovative solutions to address rising demands on agricultural systems, and the exponential growth of the field of citizen science.

History, Epidemic Evolution, and Model Burn-In for a Network of Annual Invasion: Soybean Rust.

Ecological history may be an important driver of epidemics and disease emergence. We evaluated the role of history and two related concepts, the evolution of epidemics and the burn-in period required for fitting a model to epidemic observations, for the U.S. soybean rust epidemic (caused by Phakopsora pachyrhizi). This disease allows evaluation of replicate epidemics because the pathogen reinvades the United States each year. We used a new maximum likelihood estimation approach for fitting the network model based on observed U.S. epidemics. We evaluated the model burn-in period by comparing model fit based on each combination of other years of observation. When the miss error rates were weighted by 0.9 and false alarm error rates by 0.1, the mean error rate did decline, for most years, as more years were used to construct models. Models based on observations in years closer in time to the season being estimated gave lower miss error rates for later epidemic years. The weighted mean error rate was lower in backcasting than in forecasting, reflecting how the epidemic had evolved. Ongoing epidemic evolution, and potential model failure, can occur because of changes in climate, host resistance and spatial patterns, or pathogen evolution.

A Coordinated Effort to Manage Soybean Rust in North America: A Success Story in Soybean Disease Monitoring.

Existing crop monitoring programs determine the incidence and distribution of plant diseases and pathogens and assess the damage caused within a crop production region. These programs have traditionally used observed or predicted disease and pathogen data and environmental information to prescribe management practices that minimize crop loss. Monitoring programs are especially important for crops with broad geographic distribution or for diseases that can cause rapid and great economic losses. Successful monitoring programs have been developed for several plant diseases, including downy mildew of cucurbits, Fusarium head blight of wheat, potato late blight, and rusts of cereal crops. A recent example of a successful disease-monitoring program for an economically important crop is the soybean rust (SBR) monitoring effort within North America. SBR, caused by the fungus Phakopsora pachyrhizi, was first identified in the continental United States in November 2004. SBR causes moderate to severe yield losses globally. The fungus produces foliar lesions on soybean (Glycine max) and other legume hosts. P. pachyrhizi diverts nutrients from the host to its own growth and reproduction. The lesions also reduce photosynthetic area. Uredinia rupture the host epidermis and diminish stomatal regulation of transpiration to cause tissue desiccation and premature defoliation. Severe soybean yield losses can occur if plants defoliate during the mid-reproductive growth stages. The rapid response to the threat of SBR in North America resulted in an unprecedented amount of information dissemination and the development of a real-time, publicly available monitoring and prediction system known as the Soybean Rust-Pest Information Platform for Extension and Education (SBR-PIPE). The objectives of this article are (i) to highlight the successful response effort to SBR in North America, and (ii) to introduce researchers to the quantity and type of data generated by SBR-PIPE. Data from this system may now be used to answer questions about the biology, ecology, and epidemiology of an important pathogen and disease of soybean.

Predicting Soybean Rust Incursions into the North American Continental Interior Using Crop Monitoring, Spore Trapping, and Aerobiological Modeling.

Between 2005 and 2009, millions of U.S. and Canadian soybean acres that would have received fungicide application remained untreated for soybean rust due to information disseminated through the Integrated Pest Management Pest Information Platform for Extension and Education (ipmPIPE), increasing North American producers' profits by hundreds of millions of dollars each year. The results of our analysis of Phakopsora pachyrhizi urediniospores in rain collections, aerobiology model output, and observations of soybean rust spread in 2007 and 2008 show a strong correspondence between spore collections and model predictions for the continental interior of North America, where soybean is an important crop. The analysis suggests that control practices based on up-to-date maps of soybean rust observations and associated commentary from Extension Specialists delivered by the ipmPIPE may have suppressed the number and strength of inoculum source areas in the southern states and retarded the northward progress of seasonal soybean rust incursions into continental North America. The analysis further indicates that spore trapping and aerobiological modeling can reduce our reliance on the costly Sentinel Plot Network while maintaining the effectiveness of the ipmPIPE system for soybean rust management.

Removal of Wet Deposited Phakopsora pachyrhizi Urediniospores from Soybean Leaves by Subsequent Rainfall.

Urediniospores of Phakopsora pachyrhizi, the soybean rust fungus, have a high probability of being removed from a soybean leaf by water runoff associated with subsequent rainfall after wet deposition. The effects of rainfall intensity, subsequent spore-free rainfall duration, and soybean leaf sample height on uredinia density were used to evaluate the retention of urediniospores on soybean leaf tissue. Rainfall simulations of 45 and 85 mm/h were conducted on potted soybean plants that were inoculated with 2 min of urediniospore-injected simulated rainfall and exposed to 0, 1, and 30 min of subsequent spore-free rainfall. Urediniospore retention was estimated using uredinia density values obtained from a detached leaf bioassay for the sample heights of soil level, mid-canopy, and upper-canopy. Soil level leaflets inoculated with the 45 mm/h rainfall intensity treatment had a higher (P < 0.01) mean number of uredinia/cm2 than the 85 mm/h treatment, even though they were inoculated with approximately 40% fewer urediniospores. Subsequent spore-free rainfall reduced (P < 0.01) uredinia density by as much as 38 and 91% for the 1- and 30-min durations, respectively. The relationship between uredinia density proportion and depth of rainfall was best fit using an inverse power empirical model. Our results indicate that a majority of the wet deposited P. pachyrhizi urediniospores would be removed from soybean leaf surfaces by subsequent rainfall, but sufficient percentages of spores (10 to 25%) will likely remain on the leaf tissue long enough to germinate and infect during heavy summer rains lasting ≥30 min.

The Effect of Solar Irradiance on the Mortality of Phakopsora pachyrhizi Urediniospores.

Soybean rust, caused by Phakopsora pachyrhizi, may be the most important foliar disease of soybean. Within the last 10 years, the fungus has moved to many new geographical locations via spread of airborne urediniospores. The objective of this study was to determine the relationship between urediniospore viability and exposure to solar radiation. Urediniospores of P. pachyrhizi were exposed in Capitán Miranda, Paraguay, to determine the deleterious effects of sunlight. Concomitant total solar (0.285 to 2.8 µm) and ultraviolet (0.295 to 0.385 µm) irradiance measurements were used to predict urediniospore germination. Urediniospores exposed to doses of solar and ultraviolet (UV) radiation ≥27.3 MJ/m2 and ≥1.2 MJ/m2, respectively, did not germinate. The proportions of urediniospores that germinated, normalized with respect to the germination proportion for unexposed urediniospores from the same collections, were a linear function of solar irradiance (R2 = 0.83). UV measurements predicted normalized germination proportions equally well. Results of inoculation experiments with exposed P. pachyrhizi urediniospores supported the results of the germination trials, although the effects of moderate levels of irradiance varied. The relationship between urediniospore viability and exposure to solar radiation has been incorporated into the U.S. Department of Agriculture's soybean rust aerobiological model that provides North American soybean growers with decision support for managing soybean rust.

Distribution of Leaf-Feeding Beetles and Bean pod mottle virus (BPMV) in Illinois and Transmission of BPMV in Soybean.

Bean leaf beetles (BLB; Cerotoma trifurcata) were collected in soybean (Glycine max) fields in 58 and 99 Illinois counties surveyed during the 2000 and 2001 growing seasons, respectively. In 2000, BLB counts were highest in the central portion of the state. BLB counts were lower the following year, but were more uniformly distributed throughout the state. BLB tested positive for Bean pod mottle virus (BPMV) in 37 of 41 counties assayed in 2000. In 2001, BLB tested positive for BPMV in 86 of 99 counties sampled. In 2000 and 2001, western corn rootworm (WCR; Diabrotica virgifera virgifera) adults were abundant in soybean fields only in east central Illinois. WCR adults tested positive for BPMV in 21 of 21 east central Illinois counties in 2000 and 20 of 24 sampled in 2001. BPMV was detected in soybean plants in 38 of 46 counties sampled in 2000. Field-collected WCR adults transmitted BPMV to potted soybean plants at low rates either directly from BPMV-infected soybean fields or with prior feeding on BPMV-infected plants. This is the first report of the distribution of BLB, WCR adults, and BPMV in Illinois and of BPMV transmission by adult WCR.

Mitigating epidemics caused by non-persistently transmitted aphid-borne viruses: the role of the pliant environment.

Using Soybean mosaic virus as a model system, the rate, magnitude and timing of epidemics caused by non-persistently transmitted, aphid-borne viruses are examined under various field conditions. Emphasis is placed on the behavioural responses of vectors to environmental cues, although all three biotic components (host plant, vector and virus) are considered. Both single and double manipulations of the cropping system environment are explored using a computer model developed earlier by Ruesink and Irwin (Plant Virus Epidemics: Monitoring, Modelling and Predicting Outbreaks.

Intonation and dialog context as constraints for speech recognition.

This paper describes a way of using intonation and dialog context to improve the performance of an automatic speech recognition (ASR) system. Our experiments were run on the DCIEM Maptask corpus, a corpus of spontaneous task-oriented dialog speech. This corpus has been tagged according to a dialog analysis scheme that assigns each utterance to one of 12 "move types," such as "acknowledge," "query-yes/no" or "instruct." Most ASR systems use a bigram language model to constrain the possible sequences of words that might be recognized. Here we use a separate bigram language model for each move type. We show that when the "correct" move-specific language model is used for each utterance in the test set, the word error rate of the recognizer drops. Of course when the recognizer is run on previously unseen data, it cannot know in advance what move type the speaker has just produced. To determine the move type we use an intonation model combined with a dialog model that puts constraints on possible sequences of move types, as well as the speech recognizer likelihoods for the different move-specific models. In the full recognition system, the combination of automatic move type recognition with the move specific language models reduces the overall word error rate by a small but significant amount when compared with a baseline system that does not take intonation or dialog acts into account. Interestingly, the word error improvement is restricted to "initiating" move types, where word recognition is important. In "response" move types, where the important information is conveyed by the move type itself--for example, positive versus negative response--there is no word error improvement, but recognition of the response types themselves is good. The paper discusses the intonation model, the language models, and the dialog model in detail and describes the architecture in which they are combined.

Children's deictic and nondeictic interpretations of the spatial locatives in front of and behind.

A younger group (mean age 5; 10) and an older group (mean age 9; 11) of children were asked to place a toy man IN FRONT OF or BEHIND a car which either faced away from them or faced sideways. When the observer's viewpoint was not emphasized in the instructions, or was merely implied, both age groups responded in a nondeictic way. When the viewpoint of the observer was an essential component of the task and the spatial term was omitted, both age groups selected a deictic location. However, when the viewpoint was emphasized but when the spatial term was also included the responses were split between deictic and nondeictic locations.