As the effects of climate change continue to intensify, non-native species are becoming more prevalent in estuarine ecosystems. This has implications for the taxonomic and functional diversity of fish communities. Historically, biodiversity has been a synonym of taxonomic diversity, however this approach often fails to provide accurate insights on ecosystem functioning and resilience. To better understand how climate change is impacting fishes and their traits’ composition, a long-term dataset from Minho Estuary (NW Iberian Peninsula) fish assemblage was analyzed. The results suggest that climate change and extreme weather events altered the prevailing trait modalities of fishes, which led to the overall decrease in functional diversity of the fish assemblage over the course of a decade. This decrease is associated to the loss of some trait modalities that are exclusively found in native species. On the other hand, the invasive species added novel traits associated with the conditions of high temperatures and low precipitation regime currently observed in the studied area. Our results highlight that the shift in the presence and dominance of some traits is directly influenced by climatic changes. Also, despite the addition of novel modalities by the invasive species, the fish assemblage is now less functional and taxonomically diverse than previously.

Graphical abstract

biodiversity, biological invasions, climatic events, ecosystem functioning, native species, traits

Climate change is one of the biggest threats to biodiversity currently (IPBES 2019; Reid et al. 2019), and nowadays many taxonomic groups and ecosystems around the world have already been affected (Alan Pounds et al. 2006; Moritz et al. 2008; Jarić et al. 2019). As global temperatures rise and weather patterns shift, ecosystems across the globe are undergoing drastic changes in their stability and functioning (Markham 1996; Walther et al. 2002; Parmesan and Yohe 2003). This can have a significant impact on the functional diversity of an ecosystem, as climate change can either favor certain species or cause the local extinction of others (Thuiller et al. 2006). The effects of climate change on functional diversity vary depending on the types of species present in a particular ecosystem. For example, species that are heat-tolerant may spread and flourish during periods of warmer climate; this is expected, for example, for the largemouth bass (Micropterus salmoides) in Iberian Peninsula (Bae et al. 2018). Other species, however, may be negatively impacted by shifts in temperature, which may translate into changes in species physiology, phenology, behavior, and geographic range (Robinson et al. 2009; Sorte et al. 2010; Hauser et al. 2018; Howard et al. 2020); this is the case for several tropical and subtropical fish species that have reach the northern Gulf of Mexico (Fodrie et al. 2010). Additionally, some species may require certain conditions for foraging or nesting, and climate change may limit their access to these resources, thus preventing them from obtaining enough food and reproducing normally (Segev et al. 2014; Descamps et al. 2017). These factors collectively contribute to changes in functional diversity within ecosystems, leading to disruptions in populations and communities that vary depending on the compatibility between a species’ ecological traits and the prevailing climatic conditions.

Historically, biodiversity has been associated to taxonomic diversity (Cardoso et al. 2014). However, this approach many times failed to provide insights into ecosystem functioning. Therefore, the use of traits and functional diversity indices alongside with taxonomic diversity provide a more holistic understanding about biodiversity (Hulme and Bernard-Verdier 2018). Both types of diversity metrics are important when evaluating the biodiversity of a given ecosystem (Villéger et al. 2010; Moore 2013; Teittinen and Virta 2021). In fact, due to the importance of functional diversity to biodiversity assessments, the number of scientific manuscripts integrating functional diversity into the ecological assessments has been increasing exponentially in recent years (Palacio et al. 2022). In general, ecosystems with high levels of both taxonomic and functional diversity are more stable and resilient to disturbances than ecosystems with low levels of diversity (Walker et al. 1999; Cadotte et al. 2011). By looking at the range of functions that different species perform in the ecosystem, the functional diversity indices provide a more in depth assessment of the ecosystem’s overall condition. This is because a greater variety of functions creates greater redundancy within the system, meaning that if one species is lost, there are others that can perform its role in the ecosystem (Biggs et al. 2020).

On the other hand, climate change plays an important role in the establishment and spread of invasive species, a phenomenon that is widely recognized (Stachowicz et al. 2002). Currently, biological invasions are one of the most important topics in ecology (Anderson et al. 2021). The impacts of biological invasions can be extensive and often detrimental to native ecosystems (Pyšek et al. 2020). Invasive species can disrupt food webs (Wainright et al. 2021), alter habitats (Crooks 2002; Guy-Haim et al. 2018), displace native species (Catford et al. 2018), cause biodiversity loss (Pyšek et al. 2020), alter ecosystem functioning (Haubrock et al. 2021), and lead to significant social and economic impacts (Simberloff et al. 2013; Diagne et al. 2020). Invasive species can also out-compete native species for resources (Catford et al. 2018; Ferreira-Rodríguez et al. 2018), leading to a decline in diversity (Mollot et al. 2017; Williams-Subiza and Epele 2021) that can affect the entire ecosystem and affect various taxonomic groups that are directly or indirectly linked to them (Crooks 2002; Guy-Haim et al. 2018; Goedknegt et al. 2020; Vivó-Pons et al. 2020). Additionally, climate change is likely to exacerbate the impacts of biological invasions (Rahel and Olden 2008; Diez et al. 2012; Bellard et al. 2013), as rising temperatures and changes in precipitation regimes create new opportunities for the establishment of non-native species in new areas (Stachowicz et al. 2002; McKnight et al. 2021; Souza et al. 2022b).

While it is well documented that biological invasions usually have a negative impact on taxonomic diversity (Pyšek et al. 2020; Ilarri et al. 2022), few studies have addressed the effects of this phenomenon on functional diversity (but see Sîrbu et al. 2022; Renault et al. 2022). Invasive species can fill empty trait gaps in the invaded ecosystems or replace the ones occupied by native species (Loiola et al. 2018), thereby disrupting functional diversity in these ecosystems (Hatfield et al. 2022; Linares et al. 2022). While the decline in taxonomic diversity can be accompanied by the loss of certain traits, it is crucial to understand how the decline in taxonomic diversity translates into changes in functional diversity, particularly in the context of simultaneous biological invasions and climate change. Hence, it is imperative to monitor ecosystem function and functional diversity as a means to address the threats posed by biological invasions and climate change. Through a continuous tracking of these indices, ecologists can identify areas at risk of species or functional loss (Santini et al. 2017) and pinpoint regions where appropriate management or conservation efforts are required.

Among the species groups particularly vulnerable to climate change and biological invasions are estuarine fishes (Gillanders et al. 2011; Souza et al. 2018; Lauchlan and Nagelkerken 2020; Ilarri et al. 2022). Despite their adaptability to a broad range of environmental conditions, these animals still exhibit sensitivity to the abiotic changes (e.g. in salinity and temperature) (Passos et al. 2016; Souza et al. 2018). Estuarine assemblages, in fact, often exhibit significant spatial and temporal variability (Sheaves 2009; Nagelkerken et al. 2015), emphasizing the importance of long-term datasets to facilitate a comprehensive assessment of fish assemblage dynamics and the influence of environmental drivers on them. Long-term monitoring data may also contribute to advances in invasion biology. In fact, there are few studies on the temporal aspects of invasive species impacts, and most of these studies are of short duration (less than 1 year), making it difficult to develop effective invasive management strategies and implement effective conservation measures (Matsuzaki et al. 2011). In addition, the use of traits can be useful in understanding invasion patterns and predicting which native species are most likely to be vulnerable to invasive species (Matsuzaki et al. 2011).

In this way, a long-term fish assemblage monitoring was conducted in the Minho Estuary (northwestern Iberian Peninsula) from 2010 to 2019. A recent study found that the Minho Estuary fish assemblage has been impacted by climatic changes and extreme temperature (heatwaves and cold-spells) and precipitation (dry and wet) events, which resulted in a less taxonomically diverse fish community dominated by a few invasive species (Ilarri et al. 2022). However, the temporal changes in the species trait composition of the Minho Estuary fish assemblage are still not known. To address this knowledge gap, a decade’s worth of data on fish trait composition and climate (temperature and precipitation) from Minho Estuary were compiled from weekly in-situ fish sampling and satellite data. We hypothesize that climate change and extreme weather events alter the trait and functional diversity of estuarine fish communities, with most indices of functional diversity (e.g. functional divergence, dispersion, richness, evenness and RAO’s quadratic entropy) being expected to decrease, while the functional redundancy is expected to increase. This is due to the increased prevalence of invasive species, which might introduce different traits compared to those originally found in the fish community, but also contribute to the loss or decline of some traits in the fish community.

We have been monitoring the fish populations in the Minho Estuary for over a decade to better understand the effects of changing environmental conditions on biodiversity. During this period, we have observed signs of decline in both taxonomic and functional diversities, which seems to correspond to a decreasing number of native species and an increasing prevalence of invasive species. This phenomenon seems to be further influenced by changes in environmental factors such as temperature and precipitation, which appear to impact several key trait characteristics of these fishes. Overall, there has been a significant shift in fish assemblage occurring in this estuary over the past decade, which now has an almost equal contribution of native and invasive species in terms of species richness, whereas the latter dominate in terms of abundance (Ilarri et al. 2022).

Over the past few years, the Minho Estuary has witnessed a significant increase in populations of invasive species (Sousa et al. 2013; Ilarri et al. 2022). The pumpkinseed, in particular, common carp, and tench are three species that have flourished in the estuary since 2015–2016 (Ilarri et al. 2022), and their increase is likely due to changes in the prevalent environmental conditions. Cano-Barbacil et al. (2022) and Bae et al. (2018) propose temperature as a central factor in explaining the spread of invasive species in the Iberian Peninsula. They emphasize this proposition by pointing out the strong correlation between temperature and thermophilic characteristics of most invasive species, as those from the aforementioned species. Additionally, the pumpkinseed, common carp, and tench prefer slow currents (Benito et al. 2015; Avlijaš et al. 2018; Lages et al. 2021) and highly vegetated zones (Penne and Pierce 2008; Top et al. 2016; Avlijaš et al. 2018), which are likely to become more prevalent with changes in temperature (increase) and rainfall regime (decrease) over time. These three species are also potamodromous, meaning that they perform migrations in the river. In addition, they are either phytophilic or phyto-litophilic species, reproducing in areas rich in submerged vegetation and rocks. These species are also either eurytopic (common carp) or limnophilic (pumpkinseed and tench), which means that the common carp tolerates a wide range of environmental conditions, while pumpkinseed and tench are associated with slow moving waters. Interestingly, these three species also have the shape 1 classification in terms of body shape (more rounded and compacted body) and are either average (common carp) or slow swimmers (pumpkinseed and tench) (Schmidt-Kloiber and Hering 2015). The traits’ characteristics of these three invasive species seem to have been benefited in the Minho Estuary, as the decrease in precipitation and drought events have contributed to reduced river inflow and water currents in the system. Haubrock et al. (2021) also observed a significant increase over time in short-bodied species with high body depth (shape factor 1) on the Arno River (Italy). According to Vila-Gispert et al. (2005), this trait modality can be advantageous when competing with native species in slow-flow waters. On the other hand, the lower river inflow and water currents are not good for many native species with elongated body shapes that are more associated with fast-flowing waters (rheophilic) and oceanodromous or diadromous migration modalities, such as eel, shad, three-spinned stickleback, European seabass and sea trout. These species have declined sharply over time (Ilarri et al. 2022), and the traits associated with them are also disappearing from the system.

Interestingly, there was a decline in FRic, and an increase in FRed. This suggests that the fish assemblage is losing some traits and that the invasive species are not able to replace the losses of these traits. This is somewhat expected, as invasive species usually differ from native species in their life-history and ecological traits (Vila-Gispert et al. 2005). However, for the vast majority of traits analyzed in the present study, invasive species had very similar modalities to native species, with an important exception of a few traits. One of the most striking differences is probably observed in migration, where the native species have a good amount of diadromous species and almost all invasive species are potamodromous. This result highlights that climate change is indeed seriously threatening diadromous species (Limburg and Waldman 2009; Mota et al. 2016; Braga et al. 2022; de Eyto et al. 2022), putting additional pressure on this group, which is already heavily impacted (Barbarossa et al. 2020; Duarte et al. 2021; Podda et al. 2022). For diadromous species in the Minho River, in addition to climate change, additional factors such as an increasing number of dams have exacerbated their decline. The work of Azeiteiro et al. (2021) highlights a notable reduction in Allis shad populations due to the increased dam count in the river. The decline of these species is further evident in the significant shift in trait modalities related to salinity preference. In particular, our results show a decline in species that prefer marine and brackish waters, accompanied by a replacement by species that prefer freshwater environments. It is widely reported that climate change may favor marine and brackish water species in this estuary, at least in the near-term (Souza et al. 2018, 2022a). However, this may not be true for all species, as observed in this study. The sampled area is in the upper part of the estuary, where the saline intrusion is historically not so strong, ranging from 0–2.0 psu during late summer month or droughts period (Souza et al. 2013). Despite the decrease in the river inflow and the precipitation regime, it looks like that the saline intrusion is not affecting much the upper estuary, with the exception of European seabass, which in some years can reach the upper parts of the river in summer due to higher saline intrusion (Ilarri et al. 2022). On the contrary, the change in hydrologic conditions seems to favor freshwater species that prefer slow currents or standing waters (limnophilic or eurytopic), which are also invasive (common carp, goldfish, largemouth bass, pumpkinseed, and tench). This might be explained by the decreased hydrodynamics in the area, which started to attract species with affinity to slow moving freshwater.

Another important divergence in trait modality composition between native and invasive species is in the reproduction habitat. Redundancy in this trait is low, with native species preferring to spawn in rocky areas (litophilic species), while invasive species are more associated to densely vegetated areas (phytophilic species) with some rocky bottoms (phyto-litophilic) or in sandy areas (psammophilic species). The decreased rainfall and river inflow probably contributed to the growth of submerged vegetation and the accumulation of finer substrate (sand) in the area. These conditions are also likely behind the invasion success of the aquatic plant Egeria densa in the Minho Estuary, which became very abundant after 2015 (authors’ personal observation). A change in the phenology of fish species was also observed. Previously, most species had a spawning season associated with the winter season, but with the increase of invasive species in the area, there has been a change in this trait with an increase in the occurrence of species that have a summer spawning season. Fujiwara et al. (2022) also observed an increasing pattern in non-winter spawners and a decreasing trend in winter spawners when analyzing the temporal patterns of estuarine fish communities from the northwestern Gulf of Mexico. Along with the increase in summer spawning species, a reduction in the incubation period (increase in modality less or equal to seven days) was also a feature introduced by the invasive species now present in the area.

Another trait that showed important divergence between native and invasive species is the life span, with native species having a shorter life span than invasive species. This result is interesting as it is largely recognized that successful invasive species have short life spans (e.g. Jaspers et al. 2018), but this may be different for freshwater fish species in the Iberian Peninsula (e. g. Vila-Gispert et al. 2005). In this region, many of the invasive aquatic organisms arrived several centuries ago and were influenced in the past by the wishes of the rulers of society (monarchs), who deliberately introduced species from Central Europe (Clavero 2022). This important remark is necessary because the characteristics of the traits of the invasive species currently found in the studied system may not be initially selected by the environment, but by men attempting to create an ecosystem similar to that observed in Central Europe. Of the invasive fish species recorded in our study, two originate from North America (largemouth bass and pumpkinseed), while the common carp, goldfish, and tench originate from Eurasia, and the Iberian gudgeon is native to other areas of the Iberian Peninsula but not to the Minho Estuary. Another noteworthy aspect regarding invasive species in the Minho Estuary is that a significant proportion of them consists of species targeted by recreational fisheries, which, usually have different traits when compared with their native counterparts. The introduction of these species follows a meticulous selection by humans aimed at propagating certain desirable traits for angling activities, notably larger body size and a wide ecological tolerance (Alcaraz et al. 2005; Grabowska and Przybylski 2015). Unfortunately, the illegal introduction of species targeted by recreational fishermen in the Iberian Peninsula is still a problem (Clavero and Hermoso 2011).

The traits of fish species are influenced by environmental conditions and are therefore good predictors of how fish species will respond to different climate change events (Winemiller and Rose 1992; Dahlke et al. 2020). The effects of extreme weather events on fish species varies from species to species, probably related to the sensitivity of each species to the type and intensity of the event. Overall, extreme weather events had mostly strong negative effects on fish traits modalities than positive ones (i.e. decreases on the values of trait modalities were more frequent than increases). In our study, heatwaves had the greatest impact on traits compared to the other extreme events. Indeed, Barbarossa et al. (2021) suggest that increases in water temperature constitute a larger threat to freshwater fishes than changes in high and low flow conditions. The heatwaves caused a decline in trait modalities associated with higher salinity preference, reproduction in sandy habitats (Psammophilic), longer body (species with higher shape factors), average and fast swimmers, longer incubation period, low fecundity, high egg diameter, and longer larval length. On the other hand, it was observed a total benefit for short-bodied species. Our results are in part corroborated by Fujiwara et al. (2022), that suggested that fish species sensitive to changes in temperature, generally have traits associated with longer generation time, maximum length and length at maturity. In our study, traits associated with these aspects negatively responded to extreme temperature events. Interestingly, these trait modalities were also negatively correlated with the long-term effects of temperature. Therefore, the heatwave events (especially the strong ones) are possibly accelerating the speed of change in the fish community in Minho Estuary.

Regarding extreme precipitation weather events, both dry and wet events can be critical in estuarine ecosystems due to the hydrological dynamics of these systems. Although the extreme dry events correlated with a large number of species (Ilarri et al. 2022), these events seem to have a broad effect on the whole fish community, with fish traits benefiting more than being negatively affected (mainly considering cases where there was a change in abundance ≥ 70.01%). This result differs from our expectations, as we expected that these conditions have mainly negative impacts on the fish functional diversity. Normally, extreme dry events are associated with an increase in salinity and changes in other water biochemical properties (Martinho et al. 2007; Kinard et al. 2021). In this case, salinity and water quality act as abiotic filters in the fish assemblage and select fishes with traits better adapted to harsh conditions (Kinard et al. 2021). Overall, drought events were linked to an increase in the abundance of trait modalities associated with the marine environment (brackish-marine), which was expected as the decreasing water flow can lead to stronger saltwater intrusion into the upper parts of the estuary. Drought events also positively correlated with some traits modalities related to reproduction, such as incubation period, egg diameter and larval length. The favored modalities are not in the extremes of the ranges of the traits, suggesting that they might be indicative of moderate and stable environments, which also suggested that drought events probably did not cause severe stress to the fishes in Minho Estuary. On the other hand, the extreme wet events, despite of affecting a lower number of species than the extreme dry events (see Ilarri et al. 2022), they affect negatively several trait modalities. This result was also different than expected, as areas with more precipitation are normally expected to create more stable conditions than areas submitted to dry conditions. In this sense, wet events can be expected to affect the extreme modalities of traits, and to favor the moderate modalities of traits, which was not the case for several traits in this study. The extreme wet events were mostly linked to a decline in traits associated with the marine environment, such as oceanodromous and freshwater-marine modalities, which makes sense given the lower saline influence under this condition. Other traits’ modalities that were negatively correlated were pelagic, psammophilic, shape 2 and incubation period of 7–14 (intermediary modality).

Some studies indicate that changes in functional diversity are easier to detect than changes in taxonomic diversity and serve as early warning signals for threatened ecosystems. However, in this study it was possible to see the same signal in both metrics, suggesting that in the Minho Estuary the deterioration of taxonomic and functional diversity occurred simultaneously. Each functional diversity index provided a different perspective on the functional change that is occurring in the system. For example, the decrease in the FDiv index indicates that some of the most abundant species in the system nowadays have highly convergent characteristics, while FDis, FRic and FRAO tell more or less the same story, namely that the fish assemblage is losing trait richness and diversity, and particularly rapidly after 2015, a period when the dominance of a few invasive species increased significantly. The FRed index, which is a potential early warning indicator of increasing disturbances in the system (van der Linden et al. 2016), shows that the fish assemblage is becoming more functionally redundant. This result may indicate two different things: first, that some traits that were present, but not dominant are being lost; and second, that the remaining traits are more similar to each other, which may provide some resilience to the assemblage in terms of functional stability (van der Linden et al. 2016).

The findings of this study demonstrate the negative impacts of climate change and extreme weather on fish communities in estuarine ecosystems. The decline in both taxonomic and functional diversity suggests a threat to the overall balance and health of the ecosystem. These changes show no signs of slowing down, highlighting the need for immediate and effective action to mitigate environmental damage caused by climate change. Furthermore, this loss in fish diversity has implications for local cultures and economies that rely on fish as a source of food and income. It is therefore crucial to address climate change before further harm is inflicted on fish communities and the humans they support.

Conceptualization: ATS, MI. Data curation: ATS, MI. Formal analysis: ATS, MI. Investigation: ATS, CA, ED, MI. Methodology: ATS, CA, MI. Project administration: CA. Resources: CA. Software: ATS, MI. Validation: ATS, CA, ED, MI. Visualization: ATS, MI. Writing original draft: ATS, MI. Writing, review and editing: CA, ED.

The authors acknowledge support from FCT through the Strategic Funding to CIIMAR (UIDB/04423/2020, and UIDP/04423/2020), and research contracts to ED (DL57/2016/CP1344/CT0021) and MI (DL57/2016/CP1344/CT0018). This work was partly carried out in the framework of the Migra Miño – Minho project “Protection and conservation of migratory fish in the conservation of migratory fish in the international stretch of the river Minho and its tributaries”, project co-financed by the European Regional Development Fund (ERDF) through the Interreg V-A Programme, Spain through the Interreg V-A Programme, Spain-Portugal (POCTEP), 2014–2020. The authors thank the NASA Langley Research Center (LaRC) POWER Project funded through the NASA Earth Science/Applied Science Program, for providing access to the climate data. The authors would like also to thank Eduardo Martins, Mário Jorge Araújo, Catarina Braga, António Roleira, Diogo Novais, Patrício Bouça, Ana Rita Carvalho, Mafalda Fernandes for their collaboration in the fieldwork. Open access funded by Helsinki University Library. The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.