Risk Assessment and Modeling

Source tracking and transmission risk of Campylobacter spp on mixed produce crop-livestock farms using rotational grazing


Mixed crop-livestock farms (i.e. bio-diversified farms) are farms where animals and crops are raised with the goal of utilizing the products of one for the growth of the other. Mixed produce growers in northern California have diversified their production system to include a rotation of livestock in order to manage plant biomass, add manure to the soil, and increase products produced on the farm. For produce farms, re-integrating animals back into cropland provides benefits in profitability and environmental sustainability such as, reducing pests/weeds, improving of soil fertility, strengthening of farm economies, and increasing regional food security. However, these systems face challenges, including potential food safety risks and compliance with third=[arty audit criteria and newly proposed regulations in produce production. For example, auditors and regulators may discourage or prohibit raising livestock alongside vegetable crops as done in rotational grazing for crops (e.g., leafy greens, melons) known to be vulnerable to microbial contamination by fecal-born zoonotic pathogens. Specifically, raw manure from grazing animals may introduce foodborne pathogens such as Campylobacter, Salmonella, Shiga toxin-producing Escherichia coli (STEC) into fields, and these pathogens can persist in the soil for extended periods of time. Campylobacter species are commonly isolated from pasture raised livestock (ruminants and swine) and poultry. Persistence of Campylobacter spp. In bovine manure has been documented during the composting process and fresh produce has been documented as a vehicle for campylobacteriosis illnesses. Campylobacter spp have also been found in vegetable sold at Farmer’s markets, a common venue for bio-diverse farm products. In addition, the risk of Campylobacter amplification and dissemination may be increased in systems with multi-species grazing that could promote cross-species transmission. The Current National Organic Program (NOP) requires that untreated animal manure be applied at least 120 days or 90 days prior to the harvest of crops, depending on whether the edible portions come into direct or indirect contact with the treated soil. However, there is limited data defining the microbial risk or adequate waiting period after animals are grazed on a produce field. Organic farmers are currently discouraged by some third-party auditors and the Leafy Green Marketing Agreement (LGMA) from grazing animals in produce fields during the fallow period, or they must implement a strict 12-month waiting period that is not compatible with most cropping cycles in California and other regbions. Recent outbreaks and raw milk recalls due to Campylobacter contamination traced to pasture-based dairy farms in California, in addition to the high prevalence of Campylobacter in organic/pastured raised poultry carcasses and pasture farm environments, highlights the need to investigate transmission dynamics in rotational grazing systems involving multi-species. Moreover little scientific data exists regarding the risk of inter-species transmission and resulting risk to microbial contamination of fresh produce crops. The goal of this study is to investigate the prevalence of Campylobacter spp, on farms integrating livestock and produce production by rotational grazing, and source track potential movement of strains between species (cattle-poultry; swine-poultry and small ruminants-poultry).

Investigating raccoon abundance, home range and Baylisascaris procyonis prevalence in Yosemite National Park and its association with human occupancy in Yosemite Valley


Currently, there is lack of information regarding the abundance and movements of raccoon population living in Yosemite National Park. Recently, close human-raccoon interactions have been increasingly reported in Yosemite Valley, raising concerns about zoonotic disease transmission. Moreover, due to their diet habits, raccoons may pose a challenge to reintroduction programs of endangered species into the park. This project, in collaboration with the National Park Services, aims to estimate the raccoon population, home range and roundworm prevalence in Yosemite Valley, as well as to study their association with anthropogenic factors and the overlapping with areas of importance regarding conservation of aquatic endangered species. Mark-resight, GPS collaring and flotation methodology will be used to achieve such aims. Final results will offer scientific grounded information to assess whether it is necessary to apply management practices to the raccoon population in Yosemite Valley.

Epidemiological Evaluation of the Spatio-Temporal Patterns and Risk Factors Contributing to the Early Mortality Syndrome (EMS) and White Spot Syndrome (WSS) in Sinaloa, Mexico


Sinaloa is one of the Mexican states with the highest production of farmed shrimp. The appearance of the white spot (WSSV ) infection in Mexico in 1999 had a significant economic impact due to the high mortality rate. During the last decade different strategies have been implemented for production and health management by the Aquaculture Health Committees that have reduced the economic impact, however there are still factors that are contributing to the occurrence of WSSV outbreaks, which highlights the need to conduct further investigations to establish more effective control programs. Moreover, in 2013 it has been introduced a new syndrome, early mortality syndrome (EMS), whose etiology is a specific p strain of Vibrio parahaemolyticus which has been devastating the shrimp productions and, as consequence, production have reach levels of 1999 . The Committee on Aquaculture Health of Sinaloa ( CESASIN ) has detailed production and health data from the last 8 years. We propose here to conduct an intensive epidemiological analysis of this data to identify spatiotemporal patterns of disease presentation and to identify potential risk factors for diseases affecting farmed shrimp in Mexico. We will use spatial analysis and geostatistical models for such purpose.

Network analysis to identify important factors for managing zones


This WP will describe spatial and temporal dynamics of fish transportation and characterize the contact network patterns among the fish holdings based on the transportations, characterize infection-inducing contact patterns, identify the highly connected sites, and elucidate implications of the contact pattern on controlling disease spread.

Social network analysis (SNA) will be the approach used to characterize the contact pattern on the basis of fish transportation, and then identify highly connected sites and areas to be targeted for surveillance and control programs. The degree centrality (number of incoming and outgoing contacts that a site has) and closeness centrality (how closely connected each site is to all other sites within the network) will be estimated for each site in the contact network (Koschutzki et al., 2005; Dube et al 2009; Martinez-Lopez et al., 2009b). The centrality measures will subsequently be used to identify the sites at potential highest risk of receiving and/or transmitting infections within the network, and which sites, areas and time periods may play a key role for disease introduction and spread in Norwegian fish farming industry.

The working hypothesis is that “central” Norwegian fish farms (i.e., farms with high values for centrality measures) and network structure (i.e., relationship between different farms or groups of farms) have a strong influence on the vulnerability or risk on introduction and spread of diseases.

Additionally, we will use exponential random graph models (ERGM) to identify which node attributes and network structural properties influence the formation of an observed contact or, in other words, what is the probability that a movement between two sites occur given the properties and characteristics of those sites. This is useful for prediction of “future” contact patterns and ultimately, will allow the better prevention and control of diseases and the implementation of risk-reduction measures on a farming site.”

Real-time risk Assessment Platform for Evaluating the Risk of African Swine Fever (ASF) Introduction into the United States


The aim of this project is to develop a user-friendly digital platform for the real-time risk assessment of ASF introduction into the US. This platform would be scalable and could be easily adapted to evaluate other Foreign Animal Diseases (FADS). This platform will allow the gathering, update and integration of different epidemiological data for the near real-time risk assessment of ASF introduction into US through different routes. Specifically, it will provide: i) an updated risk profile (where and when) of ASF risk of introduction into the US, ii) the possibility to set up and send notifications when the risk significantly changed and iii) sensitivity analysis and scenario evaluation tools to quantify the impact of specific scenarios (e.g. the impact of change in policies or trade patterns) on the risk estimates.

Evaluating the role of direct (i.e., animal contacts) and indirect (i.e., airborne) transmission of different PRRSV genotypes within and between different swine production systems in the US


Porcine reproductive and respiratory syndrome virus (PRRS) is still one of the swine diseases responsible for large economic losses in the US, despite of preventive and control measures implemented to reduce transmission within and between production systems. However, few studies have investigated how the different PRRSV genotypes are spreading among swine systems in the US and which are the most likely transmission pathways contributing to it.

The aim of this study is to estimate the specific role of direct (i.e., animal movements) and indirect (e.g., airborne or local spread) transmission of PRRSV genotypes within and between different swine production systems in the US.

A mixed Bayesian model will be used to quantify the association between: 

  1. the pairwise genetic distance of two isolates belonging to the most frequent RFLP types (1-18-4, 1-18-2, 1-26-2, 1-4-4, etc),
  2. the spatial and temporal proximity (i.e., pairwise spatial distance and pairwise absolute difference in time between isolates),
  3. and frequency and characteristics of swine movements (i.e., social network structure: bidirectional number of incoming and outgoing shipments and number of animals moved).

Social network analysis (SNA) methods will also be used to specifically characterize the structure and characteristics of the network of animal movements. The use of SNA will allow to:

  1. identify “communities” (i.e. “groups of sites”) highly connected, with high likelihood to share PRRSV genotypes and where, for example, a “shared” or “common” PRRS control program is recommended to be implemented,
  2. characterize sites, areas and time periods at highest risk of PRRSV introduction (i.e. incoming shipments) or spread (i.e. outgoing shipments) thought animal movement and
  3. provide recommendations about the most cost-effective way to implement risk-based interventions (e.g. increase of biosecurity, vaccination, air filtering, etc.) in those sites that play a “key role” for PRRSV introduction and/or spread in order to maximize business continuity and improve PRRSV prevention and control at the local and regional level.

Moreover, we will develop a herd score index that will summarize the genetic diversity of PRRSV on site and its association with the direct (i.e. animal movement) and indirect transmission (i.e. airborne spread) potential for each site, which will provide the baseline framework for benchmarking and prioritizing interventions at an individual (i.e., high risk herds) or system level.

Methods will be also integrated in Disease BioPortal© (http://bioportal.ucdavis.edu), allowing:

  1. the near real-time update of new isolates and of trade networks into the analysis,
  2. the secure access and user-friendly visualization of the results while keeping always confidentiality (i.e. no display of sensitive/confidential information).

Methods and results are intended to provide a better understanding of how PRRSV genotypes are spreading among farms as well as to identify “critical points” (i.e. sites, areas, time periods or contacts) where attention should be focused. Specifically we will quantify the specific role that animal movements and local spread may have in disease transmission in the US swine industry and to identify high-risk herds, areas and time periods where surveillance and control strategies should be prioritized. 

Development of an early-warning system based on real-time risk assessment for the prevention and rapid control of Avian Influenza in California Poultry industry


The recent cases of highly pathogenic avian influenza (HPAI) in a commertial Turkey flock in Stanislaus country (H5N8, Jan 2015) and a commertial poultry flock (broiler chickens and ducks) in Kings county (H5N8, Feb 2015) highlights the urgent need to develop and implement solutions to protect California poultry operations (PO) against avian influenza (AI) outbreaks. The unique peculiarities of the different types of PO coexisting in California (CA) (i.e., organic vs commercial, backyard flocks, live bird markets, etc.) pose a challenge on the early detection and control of diseases such as AI which cost producers and the US millions of dollars. Mapping the occurrence of AI in wild birds and the presence of environmental and anthropogenic factors for AI occurrence has been proven useful to identify high-risk areas for poultry exposure to AI virus in countries such as China or Thailand (Gilbert et al., 2008a; Fang et al., 2013a; Gilbert et al., 2014); however, the awareness of the producers and the implementation of appropriate biosecurity and management practices on farm are key to prevent and mitigate the consequences of an AI outbreak. The aim of this project is to pilot the development an innovative early-warning system based on scientific-based risk maps, real-time notifications, on-farm risk assessments and educational tools for better prevention and control of AI outbreaks in CA. First, we will generate high-resolution AI risk maps and identify environmental, climatic and anthropogenic factors associated with AI occurrence in CA using maximum entropy ecological niche modeling. Those methods will be integrated into a web-based, dynamic, platform with capabilities to send automatic notifications to producers if changes of AI risk are detected at local or state level. Second, a self-assessment tool will allow producers to quantify the specific risk of AI virus exposure in their operations at any time given their specific location, biosecurity and management practices. Finally, we will implement workshops to increase awareness, training and responsiveness of both small-scale and large-scale producers about biosecurity practices and early detection of AI. Results of this project will built capacity, increase awareness and provide updated risk-base estimates to better prevent, detect and control AI outbreaks in CA.

Development of an Early-Warning System Based on Real-Time Risk Assessment, Producers Self-Assessment of Biosecurity Practices for the Prevention, Early Detection and Rapid Control of AI Outbreaks in the CA Poultry Industry.


The goal of this interdisciplinary, multi-institution, research-extension project within the "Critical Agricultural Research and Extension (CARE)" priority is to develop an innovative early-warning system for better prevention and control of Avian Influenza (AI) outbreaks in US poultry industry. The specific objectives are: i) produce accurate, continuosly updated, high-resolution AI risk maps and identify key factors (e.g., environmental, climatic and anthropogenic factors) associated with AI occurrence in US, ii) integrate those risk maps into a web-based platform for easy visualization and with capabilities to send automatic notifications to producers if changes of AI risk are detected at local or regional level, iii) develop a self-assessment tool where producers can quantify the risk of AIV exposure for their operations at any time given their specific location, biosecurity and management practices, iv) design, implement and test the value of outreach activities, workshops and interactive educational tools to increase awareness, training and responsiveness of small-scale and large scale producers about biosecurity practices and early detection of AI. To accomplish those goals high resolution risk maps will be produced using the cutting-edge method of maximum entropy ecological niche modeling. Data and methods will be integrated into a user-friendly web-based and mobile “app” interface to facilitate the long-term access, visualization, analysis and communication of the AI risk to producers and to provide customized recommendations and educational tools for implementing risk-mitigation measures. This work will provide valuable knowledge and operational tools for poultry producers and other stakeholders to better prevent and control AI outbreaks in the US.