Aquatic foods are crucial globally, yet existing literature on loss and waste (FLW) in this sector is largely based on case studies with significant omissions. While capture fisheries have been studied, there is a limited focus on aquaculture, now comprising around 50% of global edible aquatic food production. Many studies examine only the production or consumer ends of the supply chain, neglecting mid-supply chain losses. This research addresses these gaps by investigating aquatic food FLW in the United States, the world's largest aquatic food importer. The study aims to improve upon existing FLW estimates, particularly the FAO's 2011 estimate of 50% FLW in North America, which is outdated and based on several key assumptions: 1) FLW only originates from wild capture fisheries (ignoring aquaculture's contribution), 2) losses only occur within regional production (disregarding trade), 3) all aquatic food is sold at retail (excluding food service), and 4) consumer waste estimates rely on indirect methods rather than direct household-level measurements. The 2019 FAO report, while acknowledging these limitations, combines aquatic foods with meat and animal products, limiting the detail on aquatic-specific FLW. Therefore, this study uses primary data collection across the top ten aquatic food supply chains serving the US, supplemented with secondary data and literature, to provide updated and more accurate national estimates. The study tracks the aquatic food supply from production to consumption, encompassing both domestic and imported products from 2014 to 2018.

Existing literature on aquatic FLW primarily focuses on case studies in small-scale capture fisheries in low- and middle-income countries (LMICs), leaving significant gaps in our understanding of large-scale capture fisheries, aquaculture, and high-income country supply chains. Studies often fail to account for species-specific differences and variations in production methods, origin, and processing. Moreover, many US food waste studies group aquatic foods with meat and poultry, overlooking their unique characteristics such as perishability, diverse production methods, import patterns, distance from fishing sites to land, reliance on water quality, fragility, consumer familiarity, and characteristic odors, all factors that shape FLW patterns and influence potential solutions. The FAO’s 2011 estimate of 35% global aquatic FLW, and the higher estimates for North America, have been widely used despite their methodological shortcomings. This study addresses these limitations by employing a more comprehensive and disaggregated approach.

This study employed a mixed-methods approach, combining quantitative and qualitative data collection. The quantitative analysis focused on estimating FLW rates at each stage of the US aquatic food supply chain from 2014-2018. The study boundary started at the production stage (farm or fishery, including international sources) and ended at the point of consumption or removal from the supply chain within the US. Data were collected for the top ten species groups, representing 89% of US aquatic food consumption. For production and processing, primary data were gathered through surveys (24 producers, 20 processors) and semi-structured interviews (19 producers, 14 processors). These data were supplemented by a literature review of 19 studies providing additional loss rate data. Data on import refusals were incorporated to quantify losses from this source. For aquaculture, mortality of harvestable-sized animals was included as food loss, reflecting resource investment before death. Capture fisheries only included harvested and discarded animals as food loss, to avoid counting naturally occurring mortalities. Species-group and national loss estimates were developed using weighting factors based on the proportion of each species in the US supply. Distribution losses were calculated using survey data (5 US, 1 Canadian business), national food safety recall data, and assumed that distributors sold processed aquatic food. Retail losses were estimated using data from a Food Marketing Institute survey of US grocery store chains (90 responses), national retail sales data, and assumptions about the proportion of fresh, frozen, and canned products. Food service losses were estimated using data from 13 studies from different countries, reflecting the limited availability of specific data on US food service aquatic food waste. For consumer-level losses, a household food diary survey (n=70) was used for at-home waste estimates, with the data weighted by income to match aquatic food consumer patterns. Away-from-home waste was estimated using secondary data from food service literature. Qualitative data were collected through interviews (producers, processors, distributors, retailers, and food service) to gather insights into the causes of FLW and potential intervention strategies. The study distinguished between physical loss (food removed from the human food supply) and quality loss (edible products sold at a discount or donated). Throughout the analysis, quantities were converted to raw edible weight, and overall loss rates were calculated by summing losses at each stage and dividing by the total US supply. The study acknowledges potential sources of bias and error, including those associated with product form conversion, estimation of supply proportions, data generalizability, sampling in the distribution stage, and the use of literature values for food service and away-from-home consumption. Methods for weighting factors for production method (aquaculture vs wild capture) and origin (domestic vs imports) are elaborated in the supplementary information. The study was approved by relevant Institutional Review Boards.

The study estimated that 22.7% (0.62 MMT) of the US edible aquatic food supply (2.73 MMT) was lost or wasted annually from 2014-2018. This is substantially lower (43-55%) than previous estimates. Losses were highest at the production and consumption stages, each contributing approximately one-third of total physical loss. Production losses from imported products accounted for a significant portion (26%) of overall FLW. Specifically: * **Production:** Physical loss was 7.5% (0.21 MMT), with quality loss at 3.4% (0.09 MMT). Shrimp, catfish, salmon, canned tuna, and tilapia contributed the most to production losses. Capture fisheries had a 5.9% physical loss rate and 2.6% quality loss, while aquaculture had 8.2% and 3.8%, respectively. The difference is attributed to different assumptions on mortalities. * **Processing:** Physical loss was 1.8% (0.05 MMT), quality loss was 3.1% (0.09 MMT). Shrimp, tilapia, canned tuna, and catfish contributed the most to processing losses. Processors demonstrated a higher rate of quality loss than physical loss, suggesting effective strategies for minimizing edible food waste. * **Distribution:** Losses were minimal (1.2% physical loss, 2% quality loss). Losses included returns, unsold inventory, and recalls. * **Retail:** Physical loss was 2.9% (0.05 MMT), predominantly from fresh product discards. Strategies such as markdowns and repackaging were employed to minimize losses. * **Food Service:** Physical loss was 9.9% (0.08 MMT). Strategies included inventory management, proper storage, and cross-utilization of products. * **Consumption:** A significant portion (8.5%) of edible aquatic food was wasted at home, largely due to uneaten leftovers and over-purchasing, with additional waste (35%) coming from away-from-home meals. Common reasons for discarding included unwanted leftovers (18%) and over-purchasing (16%). Slimy appearance and odor each accounted for over 6% of wasted food. Saving money was the top motivation to reduce waste. Species-specific loss rates varied greatly, reflecting differences in production methods, fishing practices, and market demands. For example, spiny lobster in Indonesia and dagaa in East Africa showed the highest physical loss rates, while small pelagics in Indonesia and *Sardinella* spp. in Ghana had the highest quality loss rates. The Alaska pollock fishery, however, exhibited among the lowest loss rates. In aquaculture, Pangasius in Vietnam, shrimp in Vietnam and tilapia in China had high physical loss rates, while tilapia in China and Bangladesh, shrimp in Bangladesh, and several freshwater species in Bangladesh showed higher quality loss rates. The study highlights the significant contribution of imports to overall losses, demonstrating the need for international collaborations to address FLW in the US food system.

This study provides a comprehensive assessment of US aquatic FLW, revealing that current rates are significantly lower than previously estimated. The substantial contributions of production and consumption stages to overall FLW highlight opportunities for targeted interventions across these areas. The considerable variation in loss rates among species, production methods, and geographic regions underscores the need for species-specific strategies. The high proportion of losses associated with imported products stresses the importance of international cooperation to address FLW along global supply chains. The findings also highlight the trade-offs between optimizing freshness and improving labor efficiency, and the fluidity between physical and quality losses depending on local markets and preferences. The methodological differences in assessing losses from aquaculture versus capture fisheries, arising from how mortalities are handled, are also important considerations. The study's qualitative data provide insights into the strategies used by different actors in the supply chain to reduce FLW, including improved inventory management, staff training, and utilization of byproducts. The study emphasizes the importance of data-driven decision-making and highlights the need for continued data collection, species-specific analyses, and consideration of equity and gender issues in future research and interventions.

This study significantly refines estimates of aquatic food loss and waste in the United States, demonstrating that previous estimates were substantially overstated. The findings highlight the importance of disaggregated analyses across production, processing, and consumption stages, and for different species, geographic locations and supply chain structures. The high percentage of imported food products involved in loss and waste underscores the need for international cooperation to develop interventions. Future research should focus on developing and implementing targeted strategies to reduce FLW across the entire aquatic food system, paying attention to production methods (including improving water quality and disease management in aquaculture), optimizing processing techniques and by-product utilization, improving logistics and reducing waste in retail and food service, and educating consumers to minimize waste at home.

Several limitations should be considered when interpreting the findings. The conversion of aquatic foods to different product forms introduced bias, as did the assumptions made about edible portions. There were uncertainties in estimating the proportion of aquaculture versus capture fisheries and imported versus domestic products. The generalizability of the primary data to all production regions could be limited. The oversampling of specialty seafood wholesalers in the distribution stage may have skewed losses higher. Finally, the reliance on literature data for the food service stage introduced additional error. Future research should refine data collection methods to mitigate these limitations.