Social Media Reveals Potential Threat of Crayfish Trap to Birds

Abstract

Fishery bycatch is a significant threat to biodiversity, with birds being frequent casualties. Current research mainly focuses on seabird bycatch in large-scale marine fisheries, while bird bycatch in inland freshwater areas remains poorly understood. Crayfish traps are extensively used in China’s freshwater environments, but their ecological impacts on birds are overlooked due to monitoring difficulties. Through iEcology approaches, we collected and analyzed 146 bird bycatch incidents in crayfish traps from Chinese social media platforms between September 2010 and December 2023. The results revealed 420 identified birds from 62 species (11 orders, 24 families), predominantly omnivorous and carnivorous, while 106 individuals could not be identified. Cases were concentrated in the middle and lower reaches of the Yangtze River, showing significant positive correlations with water area ratio and aquaculture production (p < 0.001). During fishing seasons, the number of cases, species, and individuals were significantly higher (p < 0.001), though mortality rates increased in off seasons. The middle and lower reaches of the Yangtze River are main production areas of red swamp crayfish (Procambarus clarkii) and Chinese mitten crab (Eriocheir sinensis), where intensive use of crayfish traps may increase bird bycatch risk. Despite existing regulations, systematic supervision is needed to minimize ecosystem impacts.

1. Introduction

Fisheries are one of the main threats to biodiversity [1]. This threat comes not only from overfishing of target species but also from bycatch of non-target species [2]. Birds are easily entangled in fishing gear and are common non-target species in fisheries [3]. Current research on bird bycatch primarily focuses on the impact of large-scale offshore fishing on seabirds, and about one-third of all existing seabird species are threatened by fishery bycatch [4]. According to Žydelis et al., 100,000 to 200,000 seabirds drown in gillnets each year in the Baltic and North Seas alone [5]. A long-term study conducted by Pardo et al. in South Georgia found that bycatch in longline and trawl fisheries was the main cause of a 40–60% decline in the local wandering albatross (Diomedea exulans), grey-headed albatross (Thalassarche chysostoma), and black-browed albatross (Thalassarche melanophrys) populations over the past 35 years [6]. Additionally, bird bycatch also occurs in inland freshwater habitats. Black crowned cranes (Balearica pavonina) in Lake Firite, Chad, are at risk of excessive bycatch due to an influx of human populations [7]. Bycatch caused by fishing with monofilament nets is the primary cause of the population decline of the endangered Titicaca Grebe (Rollandia microptera) [8]. However, research on bird bycatch in freshwater environments is far less extensive than in marine environments [9].

As an efficient and portable fishing device, crayfish traps are commonly used in freshwater aquaculture ponds, rivers, and lakes across China. They are generally made of nylon nets and rigid frames, with conical entrances at both ends that are wider on the outside and narrower on the inside, allowing easy entry but making escape difficult (Figure 1). Additionally, the structural design of the crayfish trap is not only effective for catching target organisms but may also lead to accidental bird entry during their foraging activities. Unlike other fishing gear that requires constant human presence, these low-cost traps are often left unattended for extended periods, which significantly increases the risk of accidental captures. If birds are not rescued in time, they may be injured or die due to struggling, lack of food, and suffocation. However, the bycatch phenomenon of crayfish traps has often been ignored due to its concealment, sporadic occurrence, and the difficulty in data collection. The widespread deployment of these traps across various water bodies creates numerous potential hazard points for birds, yet their potential ecological impacts have been rarely explored.

In recent years, the continuous advancement of the internet has propelled the self-media industry into a phase of rapid growth, with large amounts of data uploaded by individual users being published on various social media platforms. This has undoubtedly provided a new method and data source for ecological research, referred to as ‘iEcology’. iEcology refers to an emerging research approach that uses online data for purposes other than their original intent to explore ecological patterns and processes [10]. Although various types of social media data have been widely used in research on biogeography [11], invasion biology [12], biobehavior [13], and community ecology [14], use in research on the distribution patterns and scale of bird bycatch in commercial and artisanal fisheries, including in inland wetlands, has not been reported to date. We used the “iEcology” method to collect reports on bird bycatch in crayfish traps from various domestic social media platforms in China. Subsequently, we analyzed the influencing factors and distribution patterns on a large scale. This approach revealed the potential ecological threat to bird communities. This research aims to provide valuable insights for future fisheries management, bird conservation and the formulation of related policies.

2. Materials and Methods

Data on bird bycatch incidents were collected from major Chinese search engines and a variety of prominent social media platforms. These platforms included Douyin (the Chinese version of TikTok, >700 million monthly active users), Bilibili (a video-sharing platform similar to YouTube, >300 million monthly active users), Tieba (an online forum comparable to Reddit), Sina Weibo (a microblogging platform akin to Twitter, ~600 million monthly active users), and WeChat public accounts (a content publishing system within WeChat, >1.3 billion monthly active users). Online news sites were also utilized as information sources.

To identify relevant reports, we conducted searches using the Chinese keywords “鸟” (bird) in combination with “地笼” (crayfish trap), “蟹笼” (crab cage), or “虾笼” (shrimp cage). After removing duplicate search results, we extracted the release time (specific to the month), release location (specific to the provincial administrative unit), bird species, individual count, and life status (alive or dead) from each report. Information screening and bird species identification were carried out by at least three collaborators with rich experience in field surveys and bird identification to reduce identification errors.

The classification and protection level of birds referred to the “A Checklist on the Classification and Distribution of the Birds of China (Fourth Edition)” [15] and the IUCN Red List of Threatened Species “https://www.iucnredlist.org/ (accessed on 19 February 2025)”. In addition, for each provincial-level administrative unit and corresponding to the year of the bycatch case, we obtained data on (1) the water area ratio, defined as the total area of inland water bodies divided by the total administrative area of the province, expressed as a percentage (%), and (2) the annual production of freshwater shrimp and crabs. These data were sourced from the website of the National Bureau of Statistics “https://www.stats.gov.cn/ (accessed on 11 February 2025)” and the official websites of provincial-level administrative units across China.

We used generalized linear models (GLMs) with Poisson distribution and the log link function to analyze two aspects: (1) temporal changes in the number of bird bycatch incidents in crayfish traps over the years and (2) the effects of water area ratio and annual yield of freshwater shrimp and crabs at provincial-level administrative units on the number of birds caught in crayfish traps (significant results were expressed as p-values). Two separate GLMs were constructed:

Model 1 (temporal trend): case~Poisson(λ)

log(λ) = β₀ + β₁ year,

(1)

where case represents the number of bycatch incidents per year (defined as one trap containing one or more birds).

Model 2 (spatial pattern): individuals~Poisson(λ)

log(λ) = β₀ + β₁ water_area_ratio + β₂ total_yield,

(2)

where individuals represent the number of individual birds caught in crayfish traps, water_area_ratio is the proportion of water area in each province, and total_yield is the total yield of freshwater shrimp and crabs (10,000 tons).

For both models, we reported effect sizes as rate ratios (exp(β)) with 95% confidence intervals, along with p-values for significant results.

Model assumptions were validated through variance heterogeneity tests (Bartlett’s test) and multicollinearity assessment using variance inflation factor (VIF). Variables with VIF values less than 5 were retained in the models, indicating no significant multicollinearity.

We divided the entire year into three periods: the main harvest period for red swamp crayfish (Procambarus clarkii) from May to August [16], the main harvest period for Chinese mitten crabs (Eriocheir sinensis) from September to December [17], and the non-harvest period from January to April. Differences in the number of cases, species, and individuals of birds caught in crayfish traps between periods were evaluated using Tukey’s multiple comparison test.

Data analysis and graph drawing were performed using R (version 4.3.3); the R packages “car” (version 3.1.3), “ggplot2” (version 3.5.2), “lme4” (version 1.1.37), and “multcomp” (version 1.4.28); and ArcGIS (version 10.8).

3. Results

Our study collected a total of 146 cases of bird bycatch for crayfish traps from September 2010 to December 2023. Among them, 41 cases came from online news sites, 40 cases from Douyin, 34 cases from Sina Weibo, 19 cases from Baidu Tieba, 7 cases from WeChat public accounts, 4 cases from Bilibili, and 1 case from the findings of our field investigation. A total of 526 bird individuals were recorded in all cases, excluding 106 unidentified individuals, the remaining 420 individuals belonged to 11 orders, 24 families, and 62 species.

3.1. Bird Species Caught by Crayfish Traps

From a taxonomic perspective, the bird species most frequently caught in crayfish traps were primarily from the families Scolopacidae, Ardeidae, Anatidae, and Rallidae, with 10, 10, 9, and 6 species, respectively. In terms of the number of individuals bycaught, the Red-billed Starling (Spodiopsar sericeus) had the largest bycatch in crayfish traps, followed by the Common Moorhen (Gallinula chloropus) and the Common Coot (Fulica atra), with numbers of 103, 59, and 20, respectively. In terms of diet, the largest number of birds caught in crayfish traps were omnivorous (255 individuals across 37 species), followed by carnivorous (155 individuals across 21 species) and herbivorous (10 individuals across 4 species). Omnivorous, carnivorous, and herbivorous birds accounted for 60.71%, 36.91%, and 2.38% of the total identifiable individuals, respectively. In terms of species diversity, they represented 59.68%, 33.87%, and 6.45% of the total species, respectively. From the perspective of species protection level, the bycatch records included several nationally protected species. Notably, eight individuals of Baer’s Pochard (Aythya baeri), a species under national level I protection in China, were recorded. Furthermore, seven individuals from six species designated under national level II protection were bycaught: the Cotton Pygmy Goose (Nettapus coromandelianus) (1 individual), Mandarin Duck (Aix galericulata) (2 individuals), Asian Dowitcher (Limnodromus semipalmatus) (1 individual), Tundra Swan (Cygnus columbianus) (1 individual), Short-eared Owl (Asio flammeus) (1 individual), and Common Kestrel (Falco tinnunculus) (1 individual). According to the IUCN Red List, the bycatch also involved species of global conservation concern. The eight individuals of Baer’s Pochard (Aythya baeri) are listed as critically endangered (CR). One individual of the Black-capped Kingfisher (Halcyon pileata), a vulnerable (VU) species, was recorded. Additionally, three near-threatened (NT) species were documented: Japanese Quail (Coturnix japonica) (1 individual), Grey-tailed Tattler (Tringa brevipes) (1 individual), and the aforementioned Asian Dowitcher (Limnodromus semipalmatus) (1 individual). Full details of all bycaught species and their conservation status are provided in Table S1.

3.2. Spatial and Temporal Characteristics of Bycatch in Crayfish Traps

From a spatial perspective, the province with the most reported cases of bird bycatch in crayfish traps was Hunan Province (30 cases), followed by Jiangsu Province (17 cases) and Zhejiang Province (11 cases). The cases were mainly concentrated in the middle and lower reaches of the Yangtze River (Figure 2, covering seven provinces and cities, namely, Shanghai, Jiangsu, Zhejiang, Anhui, Jiangxi, Hubei, and Hunan), with a total of 87 cases recorded, accounting for 59.6% of all reported cases. The middle and lower reaches of the Yangtze River also recorded the largest number of bird bycatch in crayfish traps, accounting for approximately 73.76% of the total number of birds.

Regarding the temporal distribution of cases, there was a significant positive correlation between the year of case publication and the number of bird bycatch cases in crayfish traps (p < 0.001, Figure 3a). Across different periods, both S2 and S3 showed significantly higher numbers of bird bycatch cases compared to S1 (both p < 0.001), with no significant difference between S2 and S3 (

Table 1. Results of temporal trend analysis and period comparisons for bird bycatch cases.

Analysis	Comparison	β Estimate	95% CI	p-Value
Temporal trend	Year effect	0.264	[0.211, 0.318]	<0.001 ***
Period comparison	S2 vs. S1	1.008	[0.530, 1.485]	<0.001 ***
	S3 vs. S1	0.959	[0.478, 1.440]	<0.001 ***
	S3 vs. S2	−0.049	[−0.402, 0.305]	0.960

Note: S1: non-fishing period (January to April); S2: main red swamp crayfish harvesting period (May to August); S3: main Chinese mitten crab harvesting period (September to December). *** p ≤ 0.001.

, Figure 3b).

In different periods, the number of bycatch bird species in crayfish traps followed the order S3 > S2 > S1, while the number of individual birds followed the order S2 > S3 > S1. The multiple comparison results (

Table 2. Multiple comparisons of bird bycatch numbers across different periods.

Response	Comparison	β Estimate	95% CI	p-Value
Species	S2 vs. S1	0.865	[0.388, 1.342]	0.001 **
	S3 vs. S1	0.965	[0.495, 1.435]	<0.001 ***
	S3 vs. S2	0.100	[−0.258, 0.458]	0.846
Individuals	S2 vs. S1	1.029	[0.793, 1.265]	<0.001 ***
	S3 vs. S1	0.587	[0.334, 0.840]	<0.001 ***
	S3 vs. S2	−0.442	[−0.635, −0.249]	<0.001 ***

Note: ** p ≤ 0.01 (or specifically p = 0.001 in this case); *** p < 0.001. S1, S2, and S3 refer to the non-fishing period (January to April), the main red swamp crayfish harvesting period (May to August), and the main Chinese mitten crab harvesting period (September to December), respectively.

, Figure 4a,b) indicated significant differences in the number of bycatch bird species between S1 and S2 (p < 0.01) and a significant difference between S1 and S3 (p < 0.001), while the number of individuals showed significant differences across all periods (p < 0.001).

3.3. Factors Influencing the Number of Bird Bycatch in Crayfish Traps

The results from the generalized linear model (

Table 3. Results of generalized linear model analyzing the relationship between bird bycatch numbers and provincial environmental factors. Values for variance inflation factor (VIF) are provided to assess multicollinearity.

Variable	β Estimate	95% CI	z-Value	p-Value	VIF
Proportion of water area	0.156	[0.141, 0.171]	20.900	<0.001 ***	1.059
Total yield of freshwater shrimp and crabs	0.006	[0.003, 0.008]	4.473	<0.001 ***	1.059

Note: *** p < 0.001.

, Figure 5a,b) showed that the number of bycatch birds in crayfish traps was significantly positively correlated with the water area ratio of different provinces (p < 0.001) and the annual production of freshwater shrimp and crabs (p < 0.001). However, the interaction between these two factors had no significant impact on the results.

3.4. Bird Mortality Rates in Crayfish Traps

Out of 526 bird individuals recorded in crayfish traps for which life status was assessed, 113 were found to have perished (

Table 4. Species and number of birds killed by crayfish traps from September 2010 to December 2023.

Order	Family	Species	Individual Number
Galliformes	Phasianidae	Japanese Quail (Coturnix japonica)	1
Anseriformes	Anatidae	Bean Goose (Anser fabalis)	5
		Mallard (Anas platyrhynchos)	2
		Eastern Spot-billed Duck (Anas zonorhyncha)	3
		Green-winged Teal (Anas crecca)	4
		Baer’s Pochard (Aythya baeri)	1
		Mandarin Duck (Aix galericulata)	1
Podicipediformes	Podicipedidae	Little Grebe (Tachybaptus ruficollis)	1
Gruiformes	Rallidae	Brown-cheeked Rail (Rallus indicus)	1
		Common Moorhen (Gallinula chloropus)	6
		Common Coot (Fulica atra)	8
Charadriiformes	Rostratulidae	Greater Painted Snipe (Rostratula benghalensis)	1
	Scolopacidae	Pintail Snipe (Gallinago stenura)	1
		Common Snipe (Gallinago gallinago)	8
		Spotted Redshank (Tringa erythropus)	3
		Green Sandpiper (Tringa ochropus)	3
Pelecaniformes	Ardeidae	Yellow Bittern (Ixobrychus sinensis)	1
		Striated Heron (Butorides striata)	1
		Chinese Pond Heron (Ardeola bacchus)	4
Passeriformes	Paridae	Cinereous Tit (Parus major)	1
	Sturnidae	Red-billed Starling (Spodiopsar sericeus)	1
	Motacillidae	Gray Wagtail (Motacilla cinerea)	1
		White Wagtail (Motacilla alba)	3
Unrecognizable			52

), yielding an overall mortality rate of 21.48%. Mortality rates across different periods: S1 (36.17%) > S3 (24.85%) > S2 (16.73%). Anatidae (16 individuals), Rallidae (15 individuals), and Scolopacidae (15 individuals) accounted for the majority of fatalities. Additionally, 52 carcasses were either too decomposed or photographed at insufficient quality to allow for species identification.

4. Discussion

4.1. Bird Bycatch Species in Crayfish Traps

In our study, the species composition of bird bycatch in crayfish traps showed a high proportion of insectivorous and omnivorous birds. The characteristics of crayfish traps might be related to this pattern. These traps are predominantly placed in shallow waters, and some become exposed to the surface due to water level fluctuations. When not collected promptly after exposure, these traps effectively become “ghost fishing gear” [18]. The accumulated catches and bait inside these traps might attract insects, which in turn could lure insectivorous and omnivorous birds. The Red-billed Starling, with its large population and insectivorous diet, often preys on insects near the water, making it the most common bycatch species. Further research is needed to understand the mechanisms by which exposed traps might affect bird behavior and contribute to bycatch incidents.

4.2. Spatial and Temporal Distribution of Bycatch Cases in Crayfish Traps

The number of cases, species, and individuals of birds bycaught in crayfish traps were spatially concentrated in the middle and lower reaches of the Yangtze River in China and were temporally concentrated from May to December. This pattern appears to be closely associated with the intensive farming of red swamp crayfish and Chinese mitten crab, the two most commonly farmed freshwater crustacean species in China. Crayfish traps are the primary fishing gear used for harvesting both species, and their widespread deployment creates numerous potential hazard points for birds. At the spatial level, our generalized linear model results revealed a significant positive correlation between the number of birds bycaught in crayfish traps and both the water area ratio and the production of freshwater shrimp and crabs in each provincial administrative region. The provinces in the middle and lower reaches of the Yangtze River have a higher water area ratio and freshwater shrimp and crab production compared to other regions, making them the main production and consumption areas of these crustaceans in China [19]. At the temporal level, the fishing activities follow distinct seasonal patterns, with red swamp crayfish mainly harvested from May to August and Chinese mitten crab from September to December. This intensive use of crayfish traps during fishing seasons has likely contributed to significantly higher bird bycatch rates during these periods compared to the non-fishing period (January to April). The significantly higher mortality rates observed during the non-fishing period might be related to both reduced trap checking, as traps are likely left idle during this period and the lower temperatures during winter months.

The observed positive correlation between publication year and reported bird bycatch incidents during 2010–2023 likely reflects two concurrent developments. Firstly, increased potential for bird–trap interactions due to the rapid expansion of China’s freshwater aquaculture, notably the crayfish industry. For instance, national crayfish farming output rose from 0.56 million tons in 2010 to 2.89 million tons in 2022, with farming area also significantly expanding [19,20]. This aquaculture intensification implies greater crayfish trap deployment and higher bycatch risk.

Secondly, the study period saw an explosion in China’s digital connectivity. The number of internet users grew from 457 million in late 2010 to over 1.092 billion by end-2023, with the number of mobile internet users surging from 303 million to over 1.091 billion. This growth, extending to rural areas (the number of rural internet users increased from 125 million to ~326 million) [21,22], dramatically increased the likelihood of bycatch incidents being observed and shared online.

Therefore, the upward trend in reported cases is best understood as a composite effect: a potential increase in actual bycatch incidents driven by aquaculture expansion, significantly amplified by the enhanced public capacity for online reporting. Disentangling these factors precisely remains challenging with iEcology data, necessitating caution when interpreting the trend in relation to actual versus reported bycatch frequency.

4.3. Limitations of This Study

The use of the “iEcology” method to obtain data still has certain limitations. Due to the influence of different users, regions, cultures, and time factors, its data may be biased to a certain extent [10]. Our study also encountered this issue to some extent.

Only 526 samples were obtained through the internet from 2010 to 2023, yet the actual number of birds bycaught in crayfish traps may be much higher than the reported statistic. The reasons for this are as follows: First, it may be limited by the development of communication technology. China’s self-media only became popular around 2010, resulting in a lack of earlier data; second, witnesses may have failed to post online about the birds found in crayfish traps for various reasons; third, some cases may not have been retrieved through keyword searches, or the social media platforms covered by the case sources were not comprehensive enough. This situation also existed in the study of Yan et al. [23].

The mortality rate of birds bycaught in crayfish traps may be relatively low. This may be because most witnesses prefer to post videos and text materials of successfully rescued birds.

Another limitation is this study’s specific focus on crayfish trap bycatch, which precluded a detailed comparative analysis with other sources of inland freshwater bird mortality (e.g., predation, habitat loss, pollution). While such a comparison would be valuable for a broader conservation perspective, it requires data and methodologies distinct from our iEcology-based approach and is therefore beyond the scope of this study. Future research should aim to integrate these multifaceted threats to develop a more holistic understanding and inform comprehensive conservation strategies.

4.4. Management Recommendations

At present, the Chinese government recognizes the impact of crayfish traps on fishery resources. Crayfish traps with mesh inner diameters of less than 30 mm were prohibited by the “List of Fishing Gear Prohibited for Use in Key Waters of the Yangtze River Basin” [24]. Some regions, such as Jiangsu and Hubei Provinces, have even completely banned crayfish traps, regardless of mesh size [25,26]. However, our analysis indicated that these areas remain hotspots for crayfish trap usage.

There may be four primary reasons for this persistence. First, crayfish traps are inexpensive and widely available for purchase. Second, their use is discreet, separating users from the traps and complicating detection and enforcement efforts. Third, there is a shortage of supervisors, and many areas lack regular patrols. Fourth, there is a lack of public awareness regarding the harm caused by crayfish traps and understanding of the regulations governing their use. To address these issues, we recommend that authorities enhance the supervision of crayfish trap distribution channels, particularly e-commerce platforms, within no-fishing zones, allowing only registered farmers to purchase traps. Clear signage should be used to display regulations and penalties for violations in these areas. Authorities and volunteer associations can leverage social media to disseminate knowledge about wildlife protection and regulations, employing a collaborative approach involving government, volunteers, and the public to control the use of crayfish traps. Additionally, in non-no-fishing areas, crayfish traps should be inspected regularly. When not in use or during drying periods, they should be thoroughly emptied to prevent them from becoming “ghost fishing gear”. We believe these measures will significantly reduce the incidence of bird bycatch and encourage fishermen to adopt more sustainable fishing practices.

5. Conclusions

Our study advances the understanding of freshwater bird bycatch through iEcology approaches, documenting 62 affected bird species and revealing distinct spatial–temporal patterns in crayfish trap incidents across China. The findings identify relationships between bycatch occurrences and environmental and aquaculture factors, emphasizing the urgent need for biodiversity conservation in freshwater ecosystems.

While social media data have inherent limitations for ecological studies, our findings provide a comprehensive assessment of freshwater bird bycatch in China’s inland waters. Future research should focus on systematic field monitoring and evaluation of specific mitigation measures to protect vulnerable bird species in these ecosystems.