Quantification of soya‐based feed ingredient entry from ASFV‐positive countries to the United States by ocean freight shipping and associated seaports

Abstract African swine fever virus (ASFV) can survive in soya‐based products for 30 days with T ½ ranging from 9.6 to 12.9 days in soya bean meals and soya oil cake. As the United States imports soya‐based products from several ASFV‐positive countries, knowledge of the type and quantity of these specific imports, and their ports of entry (POE), is necessary information to manage risk. Using the data from the International Trade Commission Harmonized Tariff Schedule website in conjunction with pivot tables, we analysed imports across air, land and sea POE of soya‐based products from 43 ASFV‐positive countries to the United States during 2018 and 2019. In 2018, 104,366 metric tons (MT) of soya‐based products, specifically conventional and organic soya bean meal, soya beans, soya oil cake and soya oil were imported from these countries into the United States via seaports only. The two largest suppliers were China (52.7%, 55,034 MT) and the Ukraine (42.9%, 44,775 MT). In 2019, 73,331 MT entered the United States and 54.7% (40,143 MT) came from the Ukraine and 8.4% (6,182 MT) from China. Regarding POE, 80.9%–83.2% of soya‐based imports from China entered the United States at the seaports of San Francisco, CA, and Seattle, WA, while 89.4%–100% entered from the Ukraine via the seaports of New Orleans, LA, and Charlotte, NC. Analysis of five‐year trends (2015–2019) of the volume of soya imports from China indicated reduction over time (with a noticeably sharp decrease between 2018 and 2019), and seaport utilization was consistent. In contrast, volume remained high for Ukrainian soya imports, and seaport utilization was inconsistent. Overall, this exercise introduced a new approach to collect objective data on an important risk factor, providing researchers, government officials and industry stakeholders a means to objectively identify and quantify potential channels of foreign animal disease entry into the United States.

sharp decrease between 2018 and 2019), and seaport utilization was consistent. In contrast, volume remained high for Ukrainian soya imports, and seaport utilization was inconsistent. Overall, this exercise introduced a new approach to collect objective data on an important risk factor, providing researchers, government officials and industry stakeholders a means to objectively identify and quantify potential channels of foreign animal disease entry into the United States.

| INTRODUC TI ON
As African swine fever virus (ASFV) continues to spread across Europe and Asia (Dixon et al., 2019), the United States Department of Agriculture has worked hard to identify potential risks for viral entry to the country and develop national response plans (USDA, 2020).
While the primary focus has been on the risk of illegal entry of pork products, along with travellers from ASFV-positive countries (Ito et al., 2020;Taylor et al., 2019), the possibility of ASFV entry via the importation of contaminated feed ingredients continues to gain recognition, based on a growing body of scientific evidence (Dee et al., 2018(Dee et al., , 2020Niederwerder et al., 2019;Stoian et al., 2019).
Recent publications have described the transmission of ASFV to naïve pigs following consumption of contaminated feed, along with the calculation of the minimum infectious oral dose in feed . Survival of ASFV in several feed ingredients has been documented out to at least 30 days post-inoculation using shipping models simulating movement of feed ingredients from Eastern Europe to the United States (Dee et al., 2018;Stoian et al., 2019).
A consistent observation across all these studies was the ability of ASFV to survive in soya-based products, that is conventional (high protein/low fat) soya bean meal, organic (low protein/high fat) soya bean meal and soya oil cake, with reported half-lives of 9.6, 12.9 and 12.4 days, respectively (Dee et al., 2018;Stoian et al., 2019). This information justifies the need to understand the countries of origin of these specific ingredients, the respective volumes imported and US ports of entry (POE) utilized. Access to these data would allow regulatory agencies to focus efforts and dedicate resources to a subset of critical ports, rather than the 329 US ports of entry (seaports, border crossings and airports) currently overseen by Customs and Border Protection (United States Customs & Border Protection, 2020). Therefore, the purpose of this short communication was to conduct an analytical exercise to generate this information.

| ME THODS
The exercise focused primarily on the years 2018 and 2019, but also evaluated data from 2015 to 2019. Information on the type and quantity of soya-based feed ingredients and their specific POE was obtained at the International Trade Commission Harmonized Tariff Schedule website (www.hs.usitc.gov), a publicly available website that provides a transaction of specific trade commodities between the United States and its international trading partners. In the website database, each trade commodity was identified by a specific 10digit code known as the Harmonized Tariff Schedule (HTS), which was used for determining tariff classifications for all goods imported into the United States. Each commodity was classified based on the product's name, use and the material type, resulting in over 17,000 unique classification code numbers. Importing countries selected for inclusion in the analysis were obtained from the 43 ASFV-positive countries listed on the Canadian Food Inspection Agency (CFIA) ASFV Watch List (Appendix A). These countries, spread across Asia, Africa and Europe, have been determined high-risk areas for potential ASFV contamination of feed (Barr, 2019). Specific queries on eight specific HTS codes pertaining to soya-based feed ingredients and the 43 countries were designed on the USITC website to create a comprehensive analysis which provided information on country of origin, quantity of product, year of entry and POE into the United States for each HTS code. Data were exported into Microsoft Excel and filtered into pivot tables to answer a series of questions: 1. What are the types of soya-based products that enter the United States from the 43 ASFV-positive countries?
2. Across the 43 ASFV-positive countries, where do most of the soya-based products come from?
3. What POE receive these high-risk imports? 4. Do POE for soya-based products change over time?

| RE SULTS
Upon completion of the analysis, answers to the questions were as follows: The USITC database identified eight HTS codes that pertained to soya-based feed ingredients: various types of soya beans, soya bean meal, soya oil cake and soya oil. These eight specific 10-digit HTS codes were identified as soya-based commodities with the potential to be included in swine diets (Table 1) (Table 3).  (Table 4a). Based on data from question 2, POE summaries specific for China (Table 4b) and the Ukraine (Table 4c)

| D ISCUSS I ON
As the US feed supply becomes increasingly globalized, the risk of foreign animal diseases entering the country is significantly increased,  Despite these strengths, this approach was not without limitations. Table 1 describes eight specific 10-digit HTS codes that were selected to be included in this study based on their potential to both harbour viable virus and be fed to pigs; however, each of these products does not share the same amount of risk to the US swine population. For example, of these eight specific products, only soya beans, soya oil cake and soya bean meal are significant risks in terms of both their likelihood to be fed to pigs and their documented ability to enhance survival of ASFV for extended periods. These products were also deemed high risk because they are major components of swine rations throughout the industry.
Another limitation of the approach was the lack of information on final product destination or intended use; therefore, it was not possible to determine how much of a product ultimately ends up in the domestic swine supply chain. In addition, the numbers presented in this study indicated the total volume of a specific product cleared by US Customs at POE. USITC defines these products as 'imports for consumption', intended for use and distribution across all industries and markets and did not provide any further information on final product destination or intended use; therefore, our methods could not determine how much of a product ultimately  ends up in the domestic swine supply chain. Furthermore, given the enormous interconnected web that is the modern global trade network, there remains some speculation of the true origin of trade products as they arrive on US shores. For example, countries may import products from one country, only to repackage them and export to another. Therefore, these data are limited to only the immediate importing country and it is not capable to determine complete travel histories of all products that clear US Customs.
In closing, we felt that the exercise was successful and enhanced the knowledge of the topic. We set out to answer four specific questions using a novel approach which gathered information that is important for the development of science-based feed biosecurity plans.
While we focused on soya-based products and ASFV-positive countries, this same approach could be applied to multiple foreign trade commodities, which could assist in the development of both human and animal food safety protocols. It is hoped that these efforts will continue to stimulate communication and collaboration between the feed and livestock industries, resulting in further research into the emerging concept of 'global feed biosecurity'. Ideally, current and future information regarding the risk of pathogen spread in feed will enhance the accuracy of risk assessments, drive the continual development of efficacious feed-based mitigation strategies and ultimately bring the health status in the country of origin into the forefront of philosophies regarding the global trade of feed ingredients.