A High-Resolution, US-scale Digital Similar of Interacting Livestock, Wild Birds, and Human Ecosystems with Applications for Multi-host Epidemic Spread

Abhijin Adiga, Ayush Chopra ∗, Mandy L. Wilson ∗, S. S. Ravi, Dawen Xie, Samarth Swarup, Bryan Lewis, Andrew Warren, John Barnes, Ramesh Raskar, Madhav V. Marathe · 2024

Read A High-Resolution, US-scale Digital Similar of Interacting Livestock, Wild Birds, and Human Ecosystems with Applications for Multi-host Epidemic Spread by Abhijin Adiga, Ayush Chopra ∗, Mandy L. Wilson ∗, S. S. Ravi, Dawen Xie, Samarth Swarup, Bryan Lewis, Andrew Warren, John Barnes, Ramesh Raskar, Madhav V. Marathe (2024) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2411-01386.txt.

Description

Abstract One Health issues, such as the spread of highly pathogenic avian influenza (HPAI), present significant challenges at the human-animal-environmental interface. Recent H5N1 outbreaks underscore the need for comprehensive modeling efforts that capture the complex interactions between various entities in these interconnected ecosystems.

Catalog record

CREATORAbhijin Adiga, Ayush Chopra ∗, Mandy L. Wilson ∗, S. S. Ravi, Dawen Xie, Samarth Swarup, Bryan Lewis, Andrew Warren, John Barnes, Ramesh Raskar, Madhav V. Marathe
YEAR2024
COLLECTIONPAPERS
FORMATTXT
FILE SIZE80 KB
VERSION1.0
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2411-01386

Browse the full library · About the Tower of Babel

Full text

The opening of the work, from the library's hosted text:

A High-Resolution, US-scale Digital Similar of Interacting Livestock, Wild Birds, and Human Ecosystems with Applications for Multi-host Epidemic Spread Abhijin Adiga, Ayush Chopra ∗, Mandy L. Wilson ∗, S. S. Ravi, Dawen Xie, Samarth Swarup, Bryan Lewis, Andrew Warren, John Barnes, Ramesh Raskar, Madhav V. Marathe Abstract One Health issues, such as the spread of highly pathogenic avian influenza (HPAI), present significant challenges at the human-animal-environmental interface. Recent H5N1 outbreaks underscore the need for comprehensive modeling efforts that capture the complex interactions between various entities in these interconnected ecosystems. To support such efforts, we develop a methodology to construct a synthetic spatiotemporal gridded dataset of livestock production and processing, human population, and wild birds for the contiguous United States, called a digital similar . This representation is a result of fusing diverse datasets using statistical and optimization techniques, followed by extensive verification and validation. The livestock component includes farm-level representations of four major livestock types – cattle, poultry, swine, and sheep – including further categorization into subtypes such as dairy cows, beef cows, chickens, turkeys, ducks, etc. Weekly abundance data for wild bird species identified in the transmission of avian influenza are included. Gridded distributions of the human population, along with demographic and occupational features, capture the placement of agricultural workers and the general population. We demonstrate how the digital similar can be applied to evaluate spillover risk to dairy cows and poultry from wild bird population, then validate these results using historical H5N1 incidences. The resulting subtype-specific spatiotemporal risk maps identify hotspots of high risk from H5N1 infected wild bird population to dairy cattle and poultry operations, thus guiding surveillance efforts. 1 Introduction Highly Pathogenic Avian Influenza (HPAI) poses a serious global threat to health, environment and food security. In the Americas alone, the unprecedented spread of H5N1 virus clade 2.3.4.4b has led to severe loss of wildlife [ 14 , 38 , 60 , 54 ] . In the US, the incidence among wild birds is widespread. Large-scale outbreaks in poultry and dairy cattle threaten food production [ 12 , 53 , 15 , 48 ] . There have also been several instances of zoonotic transmissions [ 16 ] through exposure to poultry and cattle, which poses a serious pandemic risk [ 37 ] . This work is motivated by the urgent need for a modeling platform to understand and respond to HPAI spread accounting for the various agents that shape and are affected by this phenomenon. In recent years, several national-scale, realistic in silico representations of populations, socioeconomic activities, and built infrastructures have been developed to study complex phenomena such as epidemiology, emergency response and food security at fine spatiotemporal resolutions [ 10 , 19 , 61 , 23 , 53 , 32 ] . Here, we refer to such synthetic datasets as digital similars . They have statistical similarity to real data, but differ from “digital twins” [ 44 , 63 , 20 , 55 ] , which are intended as precise “living” replicas of the real-world systems they represent [ 8 , 13 ] . These realistic data sets are used for risk assessment and simulation modeling, as evidenced by studies conducted during the COVID-19 pandemic to analyze the dynamics of infectious diseases [ 1 , 25 , 33 , 36 , 3 , 17 ] . The first large-scale high-resolution digital similars of socio-technical systems were developed more than two decades ago [ 23 ] . Subsequently, many products have been proposed to address problems in multiple domains, ranging from simple gridded distributions of populations with demographic attributes, to definitions of activities and interactions with built infrastructure [ 17 , 40 ] . Some of these digital similars focus on modeling the spatial distribution of livestock [ 30 , 10 , 19 , 9 , 53 , 34 ] . Of these previous efforts, those on US-scale data sets, including recent ones in the context of HPAI, focus on a single livestock type. Key data challenges stem from the need to explore and fuse diverse, often sparse, data sets, which are misaligned in format and spatial resolution. Methodological challenges in this context include the choice of appropriate objectives, assumptions and constraints in the algorithmic formulations in order to achieve realistic representations that are statistically consistent with the parent data sets (for example, composition and distribution of livestock farms). Summary of our contributions. This work presents an approach to develop a high-resolution multi-layered spatiotemporal representation of the contiguous US, henceforth referred to as the digital similar ( \mathcal{DS} ), that captures (i) the distribution of livestock populations and operations for multiple types (like cattle or poultry) and subtypes (like beef or milk cows, chickens, turkeys, etc.), (ii) associated food processing center locations, capacities, and functions, (iii) spatiotemporally-varying wild bird abundances for multiple species affected by H5N1, and (iv) human populations with demographic features and attributes capturing agricultural employment, as illustrated in Figure 1 . We leverage diverse open datasets (listed in Table 1 ), such as the Census of Agriculture, the Gridded Livestock of the World ( GLW ) dataset, eBird Status and Trends, a digital twin of the US population, and locations of livestock-related operations obtained from multiple sources. Data gaps are addressed by using a combination of statistical tools and mathematical programming. Mapping livestock populations to farms and assigning them to grid cells are cast as optimization problems and solved using integer linear programs. We perform rigorous data quality checks with reference to the source data sets, and verification and validation studies using independent data sets, including known locations of large livestock farms and H5N1 incidence reports. To the best of our knowledge, this is the first work to model multiple livestock types and subtypes on a national scale; previous works have focused on a single type of livestock (e.g., [ 53 , 10 , 34 ] ). We demonstrate its utility as a comprehensive platform for modeling and risk assessment of HPAI-like phenomena at high spatial and temporal resolutions, thus informing disease surveillance and control efforts. We assess the spillover risk from the H5N1-infected wild bird population to dairy cattle and poultry. We develop spatiotemporal risk maps for various subtypes, identifying hotspots for spillover risk and zoonosis. Figure 1 : Overview of the digital similar. A schematic of the system highlighting the various components and the dashboard through which the data is exposed is provided at the top. Two of the four livestock layers are shown. We have zoomed in on major production regions for the respective livestock. Both population density and counts of farms are depicted. Also shown are livestock and dairy processing centers. For the human population, agricultural workers are highlighted. The spatiotemporal distribution of three wild bird populations is shown in the bottom layer. Table 1 : Datasets explored to construct and validate the digital similar. Throughout the paper, each dataset will be referred to by its abbreviation. Name Abbrv. Source Description Census of Agriculture AgCensus [ 46 , 47 , 22 ] Provides location-level Ag data such as number of farms, farm sizes, crop types, fallowed status. Provides individual-level Ag data such as number of workers on a farm Gridded Livestock of the World GLW [ 57 ,

Read the complete text