Global Patterns and Drivers of Freshwater Fish Extinctions: Can We Learn From Our Losses?

Abstract

Nearly one-third of extant freshwater fish species, which account for over 50% of global fish diversity, are at risk of extinction. Despite their crucial ecological and socioeconomic importance, the extinction of freshwater fishes remains under-researched on a global scale. This is a comprehensive assessment of taxonomic, spatial, and temporal patterns of freshwater fish extinctions while identifying key extinction drivers and driver synergies. Using data from the International Union for Conservation of Nature Red List, 89 extinct freshwater fish and 11 extinct in the wild were analyzed. Taxonomic statistical analysis revealed the disproportionate impact on Cyprinidae, Leuciscidae, and Salmonidae. Estimated globally for the period 1851-2016, the modern extinction rate for freshwater fishes stands at 33.47 extinctions per million species-years (E/MSY), more than 100 times greater than the natural background extinction rate of 0.33 E/MSY. Extinction rates, when calculated per continent using the number of extinct species and the total number of species per continent, indicated that North America has the highest extinction rate (225.60 E/MSY), followed by Europe (220.26 E/MSY) and Asia (34.62 E/MSY). Although Africa is less affected, it still shows a 42-fold increase over the background rate. Bayesian modeling, reflecting cumulative species extinctions, indicated a strong association of North America and Asia with species loss (37 and 34 extinctions, respectively), a moderate one for Europe (20 extinctions) and a weak association of Africa (eight extinctions). Natural system modification, pollution, and invasive species emerged as the primary extinction drivers, often acting synergistically. Temporal trends indicate an acceleration in extinctions since the mid-20th century. This study highlights that, despite recent increases in conservation efforts, freshwater fish extinctions continue to rise, indicating the urgent need for integrated conservation strategies. Without immediate action, many species currently at risk may soon follow the same trajectory of extinction as the 100 extinct freshwater fishes of this study.

Keywords: IUCN; biodiversity; conservation; extinction rates; freshwater fishes.

FIGURE 1

Family (a) and genus (b) level distribution of freshwater fish extinctions.

FIGURE 2

Geographic distribution of extinct freshwater fish species by continent and country revealing regional extinction hotspots. Map lines delineate study areas and do not necessarily depict accepted national boundaries.

FIGURE 3

Cumulative observed extinctions of freshwater species globally and by continent with confidence intervals. Note five species from the Global and Europe trends were excluded due to imprecise last recorded dates (e.g., May trout Salmo schiefermuelleri , last recorded in the 19th century).

FIGURE 4

Global extinction driver maps per country for the four major drivers of freshwater fish extinctions, indicating regional extinction hotspots, that is, Philippines and Mexico. Extinction drivers: BRU, Biological Resource Use; INV, Invasive Non‐Native/Alien Species and Diseases; NSM, Natural System Modifications; POL, Pollution. Map lines delineate study areas and do not necessarily depict accepted national boundaries.

FIGURE 5

Bayesian heatmap of log‐mean co‐occurrence among extinction driver pairs globally. Extinction drivers: Biological Resource Use (BRU), Pollution (POL), Invasive Non‐Native/Alien Species and Diseases (INV), Natural System Modifications (NSM), Agriculture and Aquaculture (AA), Climate Change and Severe Weather (CLIM), “Unknown” (UKN), “Other” (OTH). Cell values represent the posterior mean log co‐occurrence rate (β₀), the number of observed co‐occurrences (N) for each driver pair and confidence intervals (95% CI). Color gradients indicate the strength of co‐occurrence on a logarithmic scale, with red denoting higher co‐occurrence (less negative β₀) and blue indicating lower co‐occurrence (more negative β₀).

FIGURE 6

Bayesian heatmap of log‐mean co‐occurrence among extinction driver pairs at the continental level. Extinction drivers: Biological Resource Use (BRU), Pollution (POL), Invasive Non‐Native/Alien Species and Diseases (INV), Natural System Modifications (NSM), Agriculture and Aquaculture (AA), Climate Change and Severe Weather (CLIM), “Unknown” (UKN), “Other” (OTH Cell values represent the posterior mean log co‐occurrence rate (β₀), the number of observed co‐occurrences (N) for each driver pair and confidence intervals (95% CI). Color gradients indicate the strength of co‐occurrence on a logarithmic scale, with red denoting higher co‐occurrence (less negative β₀) and blue indicating lower co‐occurrence (more negative β₀).

FIGURE 7

Temporal pattern in freshwater fish extinctions per major driver with confidence intervals (95% CI). Extinction drivers: BRU, Biological Resource Use; INV, Invasive Non‐Native/Alien Species and Diseases; NSM, Natural System Modifications; POL, Pollution.