Modeling the Impact of Crop Diseases on Global Food Security.

Plant pathology must contribute to improving food security in a safe operating space, which is shrinking as a result of declining natural resources, climate change, and the growing world population. This review analyzes the position of plant pathology in a nexus of relationships, which is mapped and where the coupled dynamics of crop growth, disease, and yield losses are modeled. We derive a hierarchy of pathogens, whereby pathogens reducing radiation interception (RI), radiation use efficiency (RUE), and harvest index increasingly impact crop yields in the approximate proportions: 1:4.5:4,700. Since the dawn of agriculture, plant breeding has targeted the harvest index as a main objective for domesticated plants. Surprisingly, the literature suggests that pathogens that reduce yields by directly damaging harvestable plant tissues have received much less attention than those that reduce RI or RUE. Ecological disease management needs to target diverse production situations and therefore must consider variation in attainable yields; this can be achieved through the reengineering of agrosystems to incorporate built-in dynamic diversity of genes, plants, and crop stands.

The global burden of pathogens and pests on major food crops.

Crop pathogens and pests reduce the yield and quality of agricultural production. They cause substantial economic losses and reduce food security at household, national and global levels. Quantitative, standardized information on crop losses is difficult to compile and compare across crops, agroecosystems and regions. Here, we report on an expert-based assessment of crop health, and provide numerical estimates of yield losses on an individual pathogen and pest basis for five major crops globally and in food security hotspots. Our results document losses associated with 137 pathogens and pests associated with wheat, rice, maize, potato and soybean worldwide. Our yield loss (range) estimates at a global level and per hotspot for wheat (21.5% (10.1-28.1%)), rice (30.0% (24.6-40.9%)), maize (22.5% (19.5-41.1%)), potato (17.2% (8.1-21.0%)) and soybean (21.4% (11.0-32.4%)) suggest that the highest losses are associated with food-deficit regions with fast-growing populations, and frequently with emerging or re-emerging pests and diseases. Our assessment highlights differences in impacts among crop pathogens and pests and among food security hotspots. This analysis contributes critical information to prioritize crop health management to improve the sustainability of agroecosystems in delivering services to societies.

Multiple-Disease System in Coffee: From Crop Loss Assessment to Sustainable Management.

Assessment of crop loss due to multiple diseases and pests (D&P) is a necessary step in designing sustainable crop management systems. Understanding the drivers of D&P development and yield loss helps identify leverage points for crop health management. Crop loss assessment is also necessary for the quantification of D&P regulation service to identify promising systems where ecosystem service provision is optimized. In perennial crops, assessment of crop losses due to D&P is difficult, as injuries can affect yield over years. In coffee, one of the first perennials in which crop loss trials were implemented, crop losses concurrent with injuries were found to be approximately 50% lower than lagged losses that originated following the death of productive branches due to D&P. Crop losses can be assessed by field trials and surveys, where yield reduction factors such as the number of productive branches that have died are quantified, and by modeling, where damage mechanisms for each injury are considered over several years.

Multiscale Phenotyping and Decision Strategies in Breeding for Resistance.

Advances in biotechnology have rendered tracking of quantitative trait loci (QTLs) a much easier task, making phenotyping, and not genotyping, the main bottleneck to integrating quantitative host plant resistance into breeding programs. The relevance of phenotyping methods is conditioned by their ability to predict the performance of a genotype at the field scale. Components of resistance represent the keystone hierarchy level between resistance expression in the field (the breeder's scale) and QTLs (the geneticist's scale). We describe approaches for upscaling processes to identify components of resistance that best predict field resistance, and for decision making for selection in breeding programs. We further highlight avenues for future research considering specific processes: disease transmission, defoliation, disease escape, polyetic processes, and interactions between components of resistance.

Variability in aggressiveness of rice blast (Magnaporthe oryzae) isolates originating from rice leaves and necks: a case of pathogen specialization?

Rice blast, caused by Magnaporthe oryzae, causes yield losses associated with injuries on leaves and necks, the latter being in general far more important than the former. Many questions remain on the relationships between leaf and neck blast, including questions related to the population biology of the pathogen. Our objective was to test the hypothesis of adaptation of M. oryzae isolates to the type of organ they infect. To that aim, the components of aggressiveness of isolates originating from leaves and necks were measured. Infection efficiency, latent period, sporulation intensity, and lesion size were measured on both leaves and necks. Univariate and multivariate analyses indicated that isolates originating from leaves were less aggressive than isolates originating from necks, when aggressiveness components were measured on leaves as well as on necks, indicating that there is no specialization within the pathogen population with respect to the type of organ infected. This result suggests that the more aggressive isolates involved in epidemics on leaves during the vegetative stage of the crop cycle have a higher probability to infect necks, and that a population shift may occur during disease transmission from leaves to necks. Implications for disease management are discussed.

Quantification of the Components of Resistance to Rice Sheath Blight Using a Detached Tiller Test Under Controlled Conditions.

Resistance of rice (Oryza sativa) to sheath blight, caused by Rhizoctonia solani, is quantitative and involves two mechanisms: physiological resistance and disease escape. The epidemiological concept of components of resistance was applied using a detached tiller method under controlled conditions, to specifically address physiological resistance to sheath blight in rice. A sclerotium was inserted below the leaf collar of individual rice tillers maintained in tubes filled with water. Different variables were measured after incubation: number of lesions, lesion length, vertical sheath colonization, presence or absence of dark margin at the edge of lesions, and survival duration of the leaf blade. Several rice varieties reported to have different levels of susceptibility to sheath blight were assessed, together with varieties that are cultivated over large areas. Although numerical differences between rice varieties were observed for all disease variables, only the number of lesions significantly differed among varieties tested in this study. The varieties Pecos and IR64 had the consistently lowest and highest disease intensities, respectively. This methodology may allow the detection of sources of resistance that specifically involve defense mechanisms. When combined with field assessment, this methodology should also enable to quantitatively assess the relative role of both mechanisms of resistance to sheath blight.

Analysis of nonlinear relationships in dual epidemics, and its application to the management of grapevine downy and powdery mildews.

Dual epidemics are defined as epidemics developing on two or several plant organs in the course of a cropping season. Agricultural pathosystems where such epidemics develop are often very important, because the harvestable part is one of the organs affected. These epidemics also are often difficult to manage, because the linkage between epidemiological components occurring on different organs is poorly understood, and because prediction of the risk toward the harvestable organs is difficult. In the case of downy mildew (DM) and powdery mildew (PM) of grapevine, nonlinear modeling and logistic regression indicated nonlinearity in the foliage-cluster relationships. Nonlinear modeling enabled the parameterization of a transmission coefficient that numerically links the two components, leaves and clusters, in DM and PM epidemics. Logistic regression analysis yielded a series of probabilistic models that enabled predicting preset levels of cluster infection risks based on DM and PM severities on the foliage at successive crop stages. The usefulness of this framework for tactical decision-making for disease control is discussed.

Genetic structure and aggressiveness of Erysiphe necator populations during grapevine powdery mildew epidemics.

Isolates of the causal ascomycete of grapevine powdery mildew, Erysiphe necator, correspond to two genetically differentiated groups (A and B) that coexist on the same host. This coexistence was analyzed by investigating temporal changes in the genetic and phenotypic structures of E. necator populations during three epidemics. Group A was present only at the start of the growing season, whereas group B was present throughout all three epidemics. Group A was less aggressive in terms of germination and infection efficiency but was more aggressive than group B in terms of the latency period, lesion diameter, and spore production. Our results are consistent with a temporal differentiation of niches, preventing recombination, and suggest an association between the disease level and the frequencies of genetic groups.

Quantification and modeling of crop losses: a review of purposes.

This review considers the cascade of events that link injuries caused by plant pathogens on crop stands to possible (quantitative and qualitative) crop losses (damage), and to the resulting economic losses. To date, much research has focused on injury control to prevent this cascade of events from occurring. However, this cascade involves a complex succession of components and processes whereby knowledge on crop loss generates entry points for management. Proposed here is a framework linking different types of knowledge on crop loss to a range of decision categories, from tactical to strategic short- or long-term. Important advances in this field are now under way, including a probabilistic treatment of the injury-damage relationship, or analyses of the sources of uncertainty attached to some components of the decision process. Management of injury profiles, rather than individual injuries, and shifts in dimensionality of crop losses are anticipated to contribute to the design of sustainable agricultural systems, and address global issues concerning food security and food safety.

An epidemiological simulation model with three scales of spatial hierarchy.

ABSTRACT An epidemiological model integrating three organizational scales of host plant populations (e.g., sites, leaves, and plants) is presented. At the lowest (site) scale, the model simulates the dynamics of vacant, latent, infectious, and removed sites. Three types of vacant sites are distinguished, depending on presence of infections at higher scales (leaf or plant). The rate of infection of each type of vacant site is computed according to ratios of autodeposition, allo-leaf-deposition, and allo-plantdeposition. At the leaf and plant scales, the rate of victimization is a function of the rate of infection of vacant sites. Sensitivity analyses showed that deposition patterns (the relative proportions of auto-, allo-leaf-, and allo-plant-depositions) and host structure (leaf size and number of leaves per plant) affected the speed of epidemics at the different scales. Model outputs conformed with results from other approaches in the case of random distribution of the disease. The model hypotheses concerning infection from autodeposited propagules, and their implications for disease epidemics, are discussed. The model can be used to derive relationships between allo-deposition ratios and disease incidences at the three scales. These relationships become simple when disease intensity is low. These relationships may be useful, e.g., to assess the potential efficiency of cultivar mixture to control epidemics. Integration of different organization scales and allo-deposition parameters enables the model to capture important features of epidemics developing in space without using explicitly spatialized variables. Such an approach could be useful to analyze other ecological processes that involve a variety of scales.

Research priorities for rice pest management in tropical Asia: a simulation analysis of yield losses and management efficiencies.

ABSTRACT A simulation study was conducted to assess the current and prospective efficiency of rice pest management and develop research priorities for lowland production situations in tropical Asia. Simulation modeling with the RICEPEST model provided the flexibility required to address varying production situations and diverse pest profiles (bacterial leaf blight, sheath blight, brown spot, leaf blast, neck blast, sheath rot, white heads, dead hearts, brown plant-hoppers, insect defoliators, and weeds). Operational definitions for management efficacy (injury reduction) and management efficiency (yield gain) were developed. This approach enabled the modeling of scenarios pertaining to different pest management strategies within the agroecological contexts of rice production and their associated pest injuries. Rice pests could be classified into two broad research priority-setting categories with respect to simulated yield losses and management efficiencies. One group, including weeds, sheath blight, and brown spot, consists of pests for which effective pest management tools need to be developed. The second group consists of leaf blast, neck blast, bacterial leaf blight, and brown plant-hoppers, for which the efficiency of current management methods is to be maintained. Simulated yield losses in future production situations indicated that a new type of rice plant with high-harvest index and high-biomass production ("New Plant Type") was more vulnerable to pests than hybrid rice. Simulations also indicated that the impact of deployment of host resistance (e.g., through genetic engineering) was much larger when targeted against sheath blight than when targeted against stem borers. Simulated yield losses for combinations of production situations and injury profiles that dominate current lowland rice production in tropical Asia ranged from 140 to 230 g m(-2). For these combinations, the simulated efficiency of current pest management methods, expressed in terms of relative yield gains, ranged from 0.38 to 0.74. Overall, the analyses indicated that 120 to 200 x 10(6) tons of grain yield are lost yearly to pests over the 87 x 10(6) ha of lowland rice in tropical Asia. This also amounts to the potential gain that future pest management strategies could achieve, if deployed.

Rice Pest Constraints in Tropical Asia: Characterization of Injury Profiles in Relation to Production Situations.

A protocol for characterizing patterns of rice cropping practices and injuries due to pathogens, insects, and weeds was developed and used in six sites in tropical Asia covering a wide range of environments where lowland rice is cultivated. The data collected in a total population of 456 individual farmers' fields were combined to site-specific weather data and analyzed using non-parametric multivariate techniques: cluster analyses with chi-square distance and correspondence analyses. The main results are: (i) patterns of cropping practices that are common across sites can be identified; (ii) conversely, injury profiles that are common across sites can be determined; (iii) patterns of cropping practices and injury profiles are strongly associated at the regional scale; (iv) weather patterns are strongly associated with patterns of cropping practices and injury profiles; (v) patterns of cropping practices and injury profiles allow for a good description of the variation in actual yield; and (vi) patterns of cropping practices and injury profiles provide a framework that accurately reflects weather variation and site diversity, and reliably accounts for variation in yield. The mean estimated yield across sites (4.12 t ha-1) corresponds to commonly cited averages in the region and indicates the potential for increased productivity with better management practices, especially an improved water supply. Injuries due to pests are secondary compared with other yield-limiting factors. Injury profiles were dominated by stem rot and sheath blight (IN1); bacterial leaf blight, plant hoppers, and leaf folder (IN2); and sheath rot, brown spot, leaf blast, and neck blast (IN3). IN1 was associated with high (mineral) fertilizer inputs, long fallow periods, low pesticide use, and good water management in (mostly) transplanted rice crops of a rice-rice rotation. IN2 was associated with direct-seeded rice crops in an intensive rice-rice rotation, where fertilizer and pesticide inputs are low and water management is poor, or where fertilizer and pesticide inputs are high and water management is adequate. IN3 corresponds to low input, labor intensive (hand weeding and transplanting) rice crops in a diverse rotation system with uncertain water supply. Weed infestation was an omnipresent constraint. This study shows the potential for developing pest management strategies that can be adapted throughout the region, rather than being site-specific.

Rice Pest Constraints in Tropical Asia: Quantification of Yield Losses Due to Rice Pests in a Range of Production Situations.

A series of experiments was conducted where a range of injuries due to rice pests (pathogens, insects, and weeds) was manipulated simultaneously with a range of production factors (fertilizer input, water supply, crop establishment method, variety) in different seasons and years. These factors were chosen to represent lowland rice production situations characterized in surveys conducted in tropical Asia and their corresponding range of attainable yield. Experiments complemented one another in exploring the response surface of rice yields to yield-limiting and yield-reducing factors. The resulting experimental data base consisted of 445 individual plots and involved 11 manipulated injuries in a range of attainable yields of 2 to 11 t ha-1. A first, nonparametric, multivariate analysis led to a hierarchy of potential injuries, from marginally (e.g., bacterial leaf blight) to extremely harmful (e.g., rice tungro disease). A second, parametric, multivariate approach resulted in a multiple regression model involving factors generated by principal component analysis on injuries that adequately described the variation in actual yield. One major finding was that some (attainable yield × injury factors) interactions significantly contributed to the description of variation in actual yield, indicating that some injuries (or their combinations) had a stronger or weaker yield-reducing effect, depending on the level of attainable yield. For instance, yield losses due to sheath blight, weed infestation, and rice tungro disease tend to increase, remain stable, and decrease, respectively, with increasing attainable yields. Back-computations using the principal component regression model estimated yield losses caused by individual injuries, using the mean injury levels in a population of farmers' fields surveyed across tropical Asia. The results indicate that sheath blight, brown spot, and leaf blast are diseases that cause important losses (between 1 and 10%) regionally. Among the insect injuries, only white heads caused by stem borers appear of relevance (2.3% yield losses). These injuries, however, do not match in importance those caused by weeds, whether outgrowing the rice crop canopy (WA) or not (WB), both types of injuries causing about 20% yield losses when considered individually. When all mean injuries were combined into one mean injury profile occurring at a regional attainable yield of 5.5 t ha-1, a mean yield loss of 37.2% was estimated, indicating that injuries were less than additive in their yield-reducing effects. Scenario analyses were conducted in a set of (production situations × injury profiles) combinations characterized from surveys in farmers' fields in tropical Asia. Depending on the scenario chosen, losses ranging from 24 to 41% were found.