The large amount of discards in the form of offal that are generated daily by industrial and artisanal fisheries and thrown into the sea constitutes one of the most important and predictable anthropogenic food subsidies (PAFS) that are being incorporated into marine ecosystems worldwide (Oro et al. 2013). Global discards generation in recent years has been estimated to be ca. 10 million t/year, with a peak of 19 million t/year in the late 1950s (Zeller et al. 2017). As a result of the high abundance and predictability of this anthropogenic food resource, together with a decrease in the natural prey availability due to industrial fisheries, fishery discards have important ecological implications at a global level for marine scavengers, including seabirds (Votier et al. 2004, Cury et al. 2011, Bicknell et al. 2013, Oro et al. 2013). Garthe et al. (1996), for instance, estimated that 5.9 million seabirds were potentially supported by fishery discards in the North Sea.

Changes in the availability and predictability of fishery discards as PAFS can help to understand the ecological role that food availability (i.e. carrying capacity) have at multiple ecological levels, including individual fitness, community dynamics and ecosystem functioning.

Seabirds constitute an excellent model for evaluating the ecological effects arising from a lack of PAFS for several reasons: seabirds are 1) one of the most important discard scavengers at a global level, 2) easy to monitor (because they breed on land) and 3) apical predators with a global distribution, which makes them suitable bioindicators of ecosystem health. The link between seabirds and fishery discards has been reviewed in several studies (Tasker et al. 2000, Arcos et al. 2008, Wagner and Boersma 2011). However, the ecological and evolutionary implications that fishery discards have as PAFS at a global level (Oro et al. 2013), as well as the current changes in fishery policies (see e.g. Borges et al. 2016), call for a new revision of the existing information and the identification of knowledge gaps.

Here we review current knowledge on the global ecological interactions between seabirds and fishery discards in order to identify the main knowledge gaps and propose new challenges for improving our understanding of the ecological role that food availability has for populations, communities and ecosystems.

We considered the information available in SCI journals (6 June 2017) on the Web of Science platform (WOS; Clarivate Analytics). We first selected articles with concomitant terms: [(Seabirds AND “Fishery Waste”) OR (Seabirds AND Discard)] in title, abstract or keywords (Search field = Topic) as a representative sample of research focusing on the effects of discards on seabirds’ ecology. A second search with concomitant terms: [(Seabirds AND Ecosystem AND Discard) OR (Seabirds AND Ecosystem AND “Fishery Waste”)] (Search field = Topic) was conducted to find studies focusing on the effects arising from seabird-discard interactions at the ecosystem level. Then, the selected studies were classified according to: 1) the species and families of seabirds interacting with fishery discards, 2) the fishing gear used and 3) the ecological parameter or effect investigated. Additionally, in order to identify the areas where ecological interactions between seabirds and fishery discards are more likely to occur (e.g. with high discard availability or high presence of scavenger seabirds) we calculated: 1) the average amount of discards (in metric tons) for each major FAO fishing area (www.fao.org) from 2004 to 2014 (raw data from www.seaaroundus.org) and 2) the main distribution areas of seabird species (identified as discard scavengers by reviewed studies)(data from www.iucn.org). We considered that the level of confluence of these species within each major FAO fishing area may vary throughout the year due to the large-scale movements of migratory species.

A total of 166 studies addressing up to 15 different ecological effects arising from seabird-discard interactions were selected and subsequently reviewed (Table 1, Supplementary Material Table S1). A total of 111 seabird species (Table S1) belonging to 14 taxonomic families (Table 1) were identified as scavengers of fishery discards. Demersal trawlers were by far the main fishing gear involving seabird-discard interactions (98% of studies). According to their attendance at fishing vessels, the most common discard scavengers were Laridae, Procellaridae and Diomedeidae (Table 1). The major FAO fishing areas with the highest discard availability per scavenger seabird species were the Northwest Pacific, the Eastern Central Atlantic and the Mediterranean and Black seas (Fig. 1).

Few studies quantified the effect of discards on seabirds’ ecology, and most (68%) focused on the amount/type of fishing discard in seabirds’ diet and on species attendance rate. In particular, we found that for the most important scavenger seabirds (Table 1) there was a lack of studies addressing potentially important ecological effects of discards in terms of food availability on: a) demographic parameters such as survival, dispersal and reproduction, b) resilience of populations against perturbations and c) individual foraging specialization (e.g. changes of predatory interactions, foraging and migratory patterns and the possible consequences of this heterogeneity for population dynamics). More specifically, the effect of discards on scavenging seabirds’ survival has only been studied in the family Laridae, and only 1% of species belonging to this family have been considered (Table 1). The effect that fishery discards have on seabirds’ breeding success has been hardly studied in the species belonging to the families Procellariidae (only 1% of species studied) and Diomedeidae (only 5% of species studied), which are two of the most important seabird families in terms of discard scavenger species (Table 1). We only found three studies in which individual differences in seabird foraging strategies were related to fishing practices (Matich et al. 2011, Wakefield et al. 2015, Votier et al. 2017). Finally, the role that fishery discards play in the resilience of populations remains unknown for 98% of seabirds identified as discards scavengers and for 99% of seabirds in general (Table 1).