Research Article |
|
Corresponding author: Paolo Tremolada ( paolo.tremolada@unimi.it ) Academic editor: Paride Balzani
© 2025 Athina Papatheodoulou, Marta Dell’Orso, Bruno Boz, Michele Spairani, Michalis Zacharia, Paolo Tremolada, Alessandro Balestrieri.
This is an open access article distributed under the terms of the Creative Commons Attribution License (CC BY 4.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Citation:
Papatheodoulou A, Dell’Orso M, Boz B, Spairani M, Zacharia M, Tremolada P, Balestrieri A (2025) Distribution and dietary habits of Lepomis gibbosus in Natura 2000 sites of Cyprus. Nature Conservation 59: 139-156. https://doi.org/10.3897/natureconservation.59.149036
|
Species introductions are widely reported as a major threat to biodiversity conservation in Natura 2000 sites. Pumpkinseed Lepomis gibbosus was introduced in Europe in the late 19th century and is currently listed as Invasive Alien Species of Union concern. To assess its potential impact on the native freshwater communities of Cyprus, we investigated pumpkinseed distribution and diet within 21 Natura 2000 sites. Pumpkinseed occurred in 15 out of the 22 surveyed reservoirs (68.2%), while it was recorded in only four of the 26 investigated rivers (15.4%), likely due to water availability, habitat constraints and physical barriers. Overall, this non-native fish was recorded in 12 Natura 2000 sites (57.1%). Diet analysis showed a preference for chironomid larvae, molluscs, and crustaceans, with evidence of ontogenetic dietary shifts. Fish were a minor prey, while there was no evidence of predation on reptiles or amphibians, suggesting that the impact of pumpkinseed on the conservation of native aquatic vertebrates may be negligible. Surveys also allowed to record several non-native fish species, including some predators, e.g. largemouth bass Micropterus salmoides, pikeperch Sander lucioperca and channel catfish Ictalurus punctatus, the diet and potential impact of which should be carefully assessed. This study highlights the importance of monitoring to mitigate the spread of non-native fish and support conservation within protected areas.
Biodiversity, diet, non-native species, protected areas, reservoirs
The Natura 2000 European Ecological Network was designed to support the conservation of vulnerable natural habitats and threatened species throughout the European Union (EU). The network includes Special Areas of Conservation (SACs) and Sites of Community Importance (SCIs), designated under the EU Habitats Directive (92/43/EEC), as well as Special Protection Areas (SPAs), designated under the EU Birds Directive (2009/147/EEC). The effectiveness of conservation measures aiming at halting the deterioration of biodiversity depends on several drivers of change; among them, “invasion by species”, that is the successful establishment of introduced species, is one of the most frequently reported “pressures” (negative drivers of change) by EU member states (
The introduction of fish species is widely recognised as a major driver of freshwater ecosystem alteration (
One of the most introduced aquatic species worldwide is the North American pumpkinseed Lepomis gibbosus (
The pumpkinseed has been reported to affect native fish (
Because of widespread availability of reservoirs, which possibly promoted its establishment and dispersal (
Cyprus has a Mediterranean climate, with a hot dry season from mid-April to mid-September; average annual rainfall is 480 mm/y, ranging from 450 mm in the south-western part of the island to nearly 1100 mm on the Troodos massif (1951 m a.s.l.) (
The study area included 21 Natura 2000 sites (30.5%) and nearby areas (Fig.
Three sampling campaigns were conducted from November 2022 to November 2023. Ten of the 22 lentic water bodies were sampled using both nets and electrofishing, while 12, small and shallow waterbodies, were sampled using only electrofishing. Seven gillnets with mesh width between 10 and 30 mm and two trammel nets (10 and 30 mm) were used. All nets were 2 m deep and varied in length from 18 m (10 mm) to 35 m (20 mm) and 50 m (30 mm) in length. The nets were cast in groups of three to sample different depths and habitat conditions (e.g.: rocky and muddy banks). They were set in the evening and retrieved early the next morning, using a small electric dinghy boat, to minimise the risk of accidental fuel pollution.
Electrofishing was carried out from the boat, using a Scubla motor-driven backpack device (ELT60 II GI_1300 Watt – 32 kW), using either direct – or pulse direct current and low – or high voltage (300–500 volts or 580–940, respectively) depending on water conductivity. Sampling was carried out along the banks, selecting, when available, potentially suitable-to-pumpkinseed stretches with aquatic vegetation and thin sediment. To uniform the sampling effort, the time-fished was set at 30 min.
Within the 21 Natura 2000 sites, 26 watercourses were selected based on previous fish sampling results (
At each sampling site where pumpkinseed were confirmed to occur, 10% of individuals (up to 10 individuals/site, as provided for in the permits we obtained by the competent authorities) were selected for stomach content analysis. Considering that the main aim of the analysis was to assess the impact of this IAS on aquatic vertebrates and macroinvertebrates, we selected the largest individuals, while juveniles (< 77.5 mm;
In the laboratory, stomachs and intestines were flushed into a Petri dish, keeping separate the relative contents, which were sorted using a stereoscopic microscope. The number of individuals of each prey item was assessed based on the number and position (left-right) of diagnostic hard parts (e.g.: mouth bones for fish, the labium and distal end of the abdomen for odonate larvae). When no diagnostic part was found, undigested remains were considered to belong to a single individual. Prey items were identified to the lowest feasible taxonomic units, using reference collections for fish and available keys for macroinvertebrates (e.g.
Results were expressed as percent frequency of occurrence (F% = number of samples containing a specific food item/total number of samples x 100), percent relative frequency of occurrence (FR%: number of occurrences of a food item/total number of occurrences x 100), percent volume (V% = total estimated volume of each food item as ingested/number of samples containing that item) and percent mean volume (Vm% = total estimated volume of each food item as ingested/ total number of samples x 100), which outlines the proportional contribution of each food item to the overall diet (
The pumpkinseed was recorded in 15 (68.2%) of the surveyed reservoirs, for a total catch of 429 individuals (Fig.
Overall composition of fish communities of the 22 waterbodies (21 reservoirs and Paralimini Lake) surveyed in 2022–2023. For those sampled by two methods, numbers are reported separately (nets – electrofishing). Site numbers (N) correspond to the numbers in Fig.
| N | Reservoir/Lake | Lepomis gibbosus | Micropterus salmoides | Rutilus rutilus | Blicca bjoerkna | Cyprinus carpio | Carassius auratus | Alburnus alburnus | Ictalurus punctatus | Sander lucioperca | Perca fluviatilis | Oerochromis aureus | Gambusia holbrooki | Anguilla anguilla |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 1–3 | Paralimni | - | - | - | - | - | - | - | - | - | - | - | 1 | - |
| 4 | Aftelotos | - | - | - | - | - | - | - | - | - | - | - | 20 | - |
| 5 | Achna | 36 | 6 | - | - | 5 | - | - | - | - | - | 16 | - | - |
| 7 | Kotsiatis | - | 2 | - | - | - | - | - | - | - | - | - | 1 | - |
| 8 | Mathiatis 1 | 25 | 8 | - | - | 1 | 5 | - | - | - | - | - | - | - |
| 9 | Mathiatis 2 | 17 | 2 | - | - | - | - | - | - | - | - | - | - | - |
| 14 | Klirou-M.-Akaki | 0–23 | 2–10 | 2 | - | - | - | - | 706–0 | - | - | - | - | - |
| 21 | Xyliatos | 26 | 9 | - | - | - | - | - | - | - | - | - | - | - |
| 25 | Lefkara | 0–18 | 4–0 | 3–0 | - | - | - | - | - | - | - | - | - | - |
| 29 | Dipotamos | 0–2 | 3–0 | 62–5 | - | 3–0 | 3–0 | - | 276–0 | 3–0 | - | - | - | - |
| 37 | Germasogeia | 14–45 | 12–9 | 50–14 | 19–0 | - | 4–0 | 222–8 | 18–0 | 39–0 | - | 1–0 | - | - |
| 45 | Kouris | 0–2 | 0–7 | 218–6 | 9–0 | 0–1 | - | 202–14 | 22–1 | 84–1 | 40–0 | - | - | - |
| 50 | Asprokremmos | 1–19 | 13–6 | 26–3 | 32–0 | 5–1 | - | 40–3 | 42–0 | 2–0 | - | - | - | - |
| 54 | Arminou | 25–81 | 34–13 | 641–1 | - | - | 22–0 | 54–5 | 9–0 | - | - | - | - | - |
| 57 | Kafizes | - | - | - | - | 1 | - | - | - | - | - | - | 3 | - |
| 59 | Kannaviou | 0–1 | - | 23–7 | - | - | - | - | 31–0 | 65–0 | - | - | - | - |
| 60 | Tsakistra | - | - | - | - | - | 5 | - | - | - | - | - | - | - |
| 67 | Pyrgos | - | - | - | - | - | - | - | - | - | - | - | 100 | - |
| 71 | Pomos | 12 | 4 | - | - | - | - | - | - | - | - | - | - | 1 |
| 74 | Agia Marina | 2 | 7 | - | - | - | - | - | - | - | - | - | - | - |
| 79 | Argaka | - | - | 340–12 | - | - | - | - | 11–4 | - | - | - | 0–1 | 0–8 |
| 83 | Evretou | 1–79 | 7–0 | 70–7 | 11–0 | - | - | 24–0 | 100–2 | 10–0 | - | - | - | - |
| Total | 41–388 | 75–85 | 1433–57 | 71–0 | 8–9 | 29–10 | 542–30 | 1215–7 | 203–1 | 40–0 | 1–16 | 0–126 | 0–9 |
In watercourses, pumpkinseed were recorded in only four rivers (15.4%) and as many sampling sites (6.6%) immediately up – or downstream a reservoir. Among permanent flow rivers, only the River Diarizos (n° 52 in Fig.
Overall, pumpkinseed were recorded in 12 Natura 2000 sites (57.1%), four SACs, two SCIs and six SPAs (see Suppl. materail 1). The mean length (± SE) of pumpkinseed was 75 ± 2 mm (N = 342; min-max = 20 – 197 mm). Most individuals (75.1%) measured less than 100 mm, while only eleven (3.2%) individuals exceeded 150 mm in length.
A total of 68 individuals were collected in the surveyed reservoirs (Table
Number (N) of fishes analysed for the assessment of pumpkinseed’s diet. The reservoirs were sampled using either nets and electrofishing (Ef), or only the latter method (cfr. methods). Site numbers (N) correspond to the numbers in Fig.
| N | Reservoir/Lake | N | Sampling method | |
|---|---|---|---|---|
| Nets + Ef | Ef | |||
| 1–3 | Paralimni | x | ||
| 4 | Aftelotos | x | ||
| 5 | Achna | 4 | x | |
| 7 | Kotsiatis | x | ||
| 8 | Mathiatis recharge weir 1 | 3 | x | |
| 9 | Mathiatis recharge weir 2 | 2 | x | |
| 14 | Klirou-Malounta – Akaki | 13 | x | |
| 21 | Xyliatos | 3 | x | |
| 25 | Lefkara | 2 | x | |
| 29 | Dipotamos | 1 | x | |
| 37 | Germasogeia | 7 | x | |
| 45 | Kouris | x | ||
| 50 | Asprokremmos | 4 | x | |
| 54 | Arminou | 15 | x | |
| 57 | Kafizes | x | ||
| 59 | Kannaviou | 1 | x | |
| 60 | Tsakistra | x | ||
| 66 | Pyrgos | x | ||
| 71 | Pomos | 2 | x | |
| 74 | Agia Marina | 1 | x | |
| 79 | Argaka | x | ||
| 83 | Evretou | 10 | x | |
| Total | 68 | 10 | 12 | |
Overall diet of the pumpkinseed (N = 91), as expressed in terms of both frequency of occurrence and volume (cfr. methods).
| Food items | F% | FR% | V% | Vm% |
|---|---|---|---|---|
| Diptera larvae | 56.0 | 29.8 | 67.2 | 37.6 |
| Tipulidae | 1.1 | 0.6 | 50.0 | 0.5 |
| Chironomidae | 53.8 | 28.7 | 68.7 | 37.0 |
| Psychodidae | 1.1 | 0.6 | 10.0 | 0.1 |
| Odonata larvae | 14.3 | 7.6 | 53.1 | 7.6 |
| Anisoptera | 13.2 | 7.0 | 47.9 | 6.3 |
| Zygoptera | 3.3 | 1.8 | 71.7 | 2.4 |
| Baetidae larvae | 2.2 | 1.2 | 25.0 | 0.5 |
| Trichoptera larvae | 2.2 | 1.2 | 25.0 | 0.5 |
| Undet. Invertebrate Larvae | 2.2 | 1.2 | 52.5 | 1.2 |
| Mollusca | 28.6 | 15.2 | 39.6 | 11.3 |
| Lumbriculidae | 1.1 | 0.6 | 100.0 | 1.1 |
| Lumbricidae | 1.1 | 0.6 | 100.0 | 1.1 |
| Coleoptera | 12.1 | 6.4 | 36.8 | 4.5 |
| Undetermined | 7.7 | 4.1 | 14.3 | 1.1 |
| Staphylinidae | 8.8 | 4.7 | 40.6 | 3.6 |
| Formicidae | 13.2 | 7.0 | 57.5 | 7.6 |
| Odonata (adults) | 3.3 | 1.8 | 30.0 | 1.0 |
| Undetermined Hymenoptera | 2.2 | 1.2 | 17.5 | 0.4 |
| Undetermined insects | 8.8 | 4.7 | 36.3 | 3.2 |
| Crustacea | 19.8 | 10.5 | 44.4 | 11.4 |
| Ostracoda | 8.8 | 4.7 | 34.4 | 3.0 |
| Procambarus clarkii | 11.0 | 5.8 | 76.0 | 8.4 |
| Fish | 8.8 | 4.7 | 70.6 | 6.2 |
| Gambusia holbrooki | 1.1 | 0.6 | 50.0 | 0.5 |
| Lepomis gibbosus | 1.1 | 0.6 | 80.0 | 0.9 |
| Rutilus rutilus | 1.1 | 0.6 | 30.0 | 0.3 |
| Undetermined fish | 5.5 | 2.9 | 81.0 | 4.5 |
| Algae | 5.5 | 2.9 | 64.0 | 3.5 |
Chi-squared test for the frequency of occurrence of the major food categories in the diets of three size-classes of Lepomis gibbosus (N = 12, 44 and 11, respectively).
| Food items | Length classes (mm) | Chi2 | P | ||
|---|---|---|---|---|---|
| 80–100 | 101–150 | 151–200 | |||
| Aquatic invertebrates | 87.0 | 64.9 | 44.0 | 9.69 | 0.008 |
| Diptera | 30.4 | 43.2 | 8.0 | 10.53 | 0.005 |
| Odonata larvae | 13.0 | 8.1 | 4.0 | 1.31 | n.s. |
| Mollusca | 26.1 | 4.1 | 24.0 | 11.9 | 0.003 |
| Procambarus clarkii | 0.0 | 1.4 | 0.0 | 0.65 | n.s. |
| Terrestrial invertebrates | 0.0 | 24.3 | 56.0 | 19.8 | <0.0001 |
| Coleoptera | 0.0 | 10.8 | 8.0 | 2.73 | n.s. |
| Formicidae | 0.0 | 4.1 | 32.0 | 20.6 | <0.0001 |
| Fish | 8.7 | 6.8 | 0.0 | 2.04 | n.s. |
Overall diet of the pumpkinseed, as assessed by the analysis of stomach and intestine contents. Percent frequency of occurrence (F%) vs. the estimated percent volume (V%) of eight major food categories (diamonds: aquatic prey; squares: terrestrial prey); isopleths connect points of equal percent mean volume (Vm%) in the diet.
Although the occurrence of introduced species in protected areas has progressively increased since the 1980s, their impact has been long underestimated by relevant authorities, hindering the implementation of effective management actions (
The relatively small number of large-sized pumpkinseed recorded may suggest two mutually non – exclusive hypotheses: 1) the species reaches sexual maturity early during development, as reported for other low-latitude populations (
While pumpkinseed populations may be regulated by inter and intra-specific predation in lentic waterbodies, their spread in tributaries appears to be hindered by both habitat requirements and presence of physical (waterfalls) or artificial barriers that affect longitudinal connectivity. Reservoirs typically overflow only during the short, rainy season, which may allow the downstream spread of non-native fish. This “drip-feeding” of river stretches from upstream waterbodies has also been reported for England, where, however, the species remains confined to lentic waters (
Given the currently negligible occurrence of the species in watercourses, the diet we recorded reflects the feeding habits of the species in lentic waterbodies. As previously reported for other Mediterranean areas of introduction, chironomid larvae formed the bulk of pumpkinseed diet (
Net-surveys allowed to assess the overall composition of the fish communities of reservoirs, with five species – channel catfish, common roach, pikeperch, largemouth bass and pumpkinseed – occurring in nearly all surveyed waterbodies. Although, considering both its diet and mean body length, the pumpkinseed is not expected to impact the conservation of other aquatic vertebrates, the diet and potential impact of other predators, e.g. largemouth bass (
The authors express their gratitude to the officers of the Water Development Department, Polina Polykarpou, Iakovos Tziortzis, Gerald Dörflinger and Constantinos Moustakas, from the Department of Fisheries and Marine Research, for their invaluable support and data provision.
The authors have declared that no competing interests exist.
The samplings were conducted following permits granted from the Water Development Department (dated 02/01/2023) and Department of Fisheries and Marine Research (dated 11/11/2022).
No use of AI was reported.
This work was funded by the Department of Forests, Ministry of Agriculture, Rural Development and Environment, through the LIFE project IP Physis (Pandoteira) – LIFE18 IPE/CY/000006.
The authors acknowledge support from the University of Milan through the APC initiative.
AP, MZ and AB contributed to research conceptualization, AP, AB, BB and MS led field activities, AB, MDO and PT carried out laboratory analyses.
Athina Papatheodoulou https://orcid.org/0000-0001-7198-5044
Paolo Tremolada https://orcid.org/0000-0003-4766-5352
Alessandro Balestrieri https://orcid.org/0000-0001-5444-2806
All of the data that support the findings of this study are available in the main text or Supplementary Information.
Sampling sites, N2K areas and habitat parameters of all sampled reservoirs
Data type: xlsx
Explanation note: The Excel file contains several sheet with the list of the 87 sampling sites, the list of the N2000 sites analyzed and the lentic waterbody characteristics.