The current change in the global climate is impacting us all by unusual floods, heavy rains and warm weather just to mention a few examples. In the oceans the impact of global climate change is perhaps especially a result of elevated temperature. This shift will continue to impact us and the majority of animal and plant populations. Data trends reveal global climate change effects ranging from increased oxygen consumption rates in fish, to changes in foraging and migrational patterns to fish community changes (Henderson, Reference Henderson2007). Future CO₂-induced climate change scenarios from global circulation models point to increasing air temperatures, with the greatest warming in the Arctic and Subarctic. Changes to the wind fields and precipitation patterns are also likely to happen. These will lead to changes in the hydrographical properties of the ocean, and the vertical stratification and circulation patterns (Drinkwater, Reference Drinkwater2005). Thus, we can expect fish populations in new habitats on a global scale to decline as well as a collapse of many fisheries species. As an example, stocks of Atlantic cod (Gadus morhua) in the Celtic and Irish Seas are likely to disappear under predicted temperature changes by the year 2100, while those in the southern North Sea and Georges Bank will decline (Drinkwater, Reference Drinkwater2005).

The shift in ecology and biogeography of marine fish species in the beginning of the 21st Century may hypothetically be a useful indicator for climate changes. However, first sightings of fish in highly unusual habitats and, or, new geographical areas may differ in relevance from case to case. Here, I have selected some recent cases which I discuss the pertinent use of as climate change indicators.

Settlement of the tropical lessepsian blue-barred parrotfish Scarus ghobban (Teleostei: Scaridae) in the eastern Mediterranean was recently observed by Bariche & Saad (Reference Bariche and Saad2005). The blue-barred parrotfish, Scarus ghobban Forsskål, 1775, is a wide-ranging Scaridae found from the eastern Pacific to the Red Sea and south to Algoa Bay in South Africa. This first observation in the Mediterranean is probably due to the migrational possibility via the Suez Canal. However, settling is highly dependent on optimum conditions, and such circumstances in a subtropical region for a tropical fish species may be due to climate change (Bariche & Saad, Reference Bariche and Saad2005).

Another lessepsian fish species, the migrant dusky spinefoot (Siganus luridus), was caught for the first time in March 2004 in the southern Tyrrhenian Sea (a part of the Mediterranean Sea off the western coast of Italy). This species migrated from the Red Sea through the Suez Canal into the Mediterranean where it was first recorded in 1956 (Castriota & Andaloro, Reference Castriota and Andaloro2005 and references therein). It then became common in Lebanon, Cyprus, the southern coast of Turkey, Rhodes, and was also recorded in Cretan waters and in the Gulf of Patras, Ionian Sea. Along the southern coast, S. luridus reached the Nile Delta, the Libyan coasts and Tunisia, where it first appeared in 1969 in the Gulf of Tunis, and later in the Gulf of Gabès in 1974, where it is now found frequently. Since the 1990s it has been observed by SCUBA divers in Maltese waters. A settled population was recently also recorded in the Pelagie Islands, Strait of Sicily, south Italy (Castriota & Andaloro, Reference Castriota and Andaloro2005).

A bit further north in the Mediterranian region, there was a recent first record of the Mediterranean parrotfish (Sparisoma cretense) in Ria Formosa, south Portugal. The Mediterranean parrotfish is a common species in the Macaronesian archipelagos (Azores, Madeira, Canaries and Cape Verde Islands), north-western coast of Africa and on the southern and eastern coasts of the Mediterranean, but it is the only species from the family Scaridae to be found in Mediterranean waters (Abecasis et al., Reference Abecasis, Bentes, Ribeiro, Machado, Oliveira, Veiga, Gonçalves and Erzini2006). The possibility that warming water occurring in the Mediterranean basin may be involved in the spreading of S. cretense outside its typical distribution range was recently also noticed in south-eastern Italy (Guidetti & Boero, 2001).

The unusual sightings of Scarus ghobban, Siganus luridus and Sparisoma cretense makes them potential candidates as indicators of climate changes, but the speculative reasonings for the unusual abundances make the observations too weak presently to be used as signs of climate changes. However, in combination with other parameters such as recent population biography assessments of the distribution of the populations as a whole the sightings may, or may not, be useful indicators.

The first record of the African lockdown (Selene dorsalis; of the tropical pelagic family Carangidae) in the Spanish waters of the Gulf of Cádiz, was made recently by Juárez et al. (Reference Juárez, Silva and Gil2006). Selene dorsalis is a schooling species, which inhabits inshore waters at depths of at least 100 m from the Cape Verde Islands and Senegal to South Africa. The presence of this species in the Gulf of Cádiz is unexpected according to Juárez et al. (Reference Juárez, Silva and Gil2006). They conclude that their record, together with other appearance records in waters off Portugal and Madeira, may indicate a displacement of the northern limit of the species distribution, which may be related to episodes of local warming-up of the water masses that converts the area into an optimal habitat for S. dorsalis. The appearance of this species in the Canary Islands and in Moroccan waters may support this hypothesis (Juárez et al., Reference Juárez, Silva and Gil2006).

In California, USA, another family member of the Carangidae, the fortune jack, Seriola peruana, a pelagic and demersal marine fish found in coastal areas of the tropical eastern Pacific Ocean, was recently reported in the southern Gulf of California for the first time, further north than its usual area of occurrence in the tropical eastern Pacific (Tavera et al., Reference Tavera, González-Acosta and de la Cruz-Agüero2005). According to local fishermen, the fortune jack is now quite common and can be found as far north as the middle of the Gulf of California. Tavera et al. (Reference Tavera, González-Acosta and de la Cruz-Agüero2005) suggest that the present record of S. peruana might signal the hypothesized pattern of oceanic warming (tropicalization phenomenon) or an indicator that they can adapt to the temperate or subtropical regime of the Gulf of California. The hypothesis that these sightings are indicators of climate changes is promising but it must, as for the first sightings, be tested before any conclusions can be made, by including several biotic and abiotic parameters related to the population of this species, before it is possible to conclude that the unusual distribution may be caused by climate changes.

In Argentinean waters Venerus et al. (Reference Venerus, Galván, Irigoyen and Gosztonyi2007) recently made the first record of the namorado sandperch, Pseudopercis numida, in a shallow rocky reef located near Pardelas Beach (42º38′S 64º16′W), Nuevo Gulf, Argentina, about 1600 km southwards of the previously reported limit. Venerus et al. (Reference Venerus, Galván, Irigoyen and Gosztonyi2007) point out that this new record represents an addition to a list of recently reported range expansions of warm-temperate reef fish species into higher latitudes within the northern Patagonian gulfs of Argentina of the dusky grouper Epinephelus marginatus (Irigoyen et al., Reference Irigoyen, Galván and Venerus2005) and the silver porgy, Diplodus argenteus (Galván et al., Reference Galván, Venerus, Irigoyen, Parma and Gosztonyi2005). Venerus et al. (Reference Venerus, Galván, Irigoyen and Gosztonyi2007) emphasizes that the mechanisms by which these species appeared in the North Patagonian Gulfs Ecoregion (sensu Sullivan Sealey & Bustamante, Reference Sullivan Sealey and Bustamante1999) are unknown. Venerus et al. (Reference Venerus, Galván, Irigoyen and Gosztonyi2007) continue with the consideration that if the new records just reflected increased sampling effort, one would expect to find both warm-temperate and temperate species incrementing their distributional ranges. Another possibility is that the new records indicate true southward expansions of warm-temperate species favoured by a possible increase in water temperature. Though no time data series of temperature exist for northern Patagonia, a statistically significant increase of 1.2°C (surface) and 0.5°C (bottom) in the mean summer water temperature was detected in San Jorge Gulf and adjacent waters (43°S–47°S) between 1995 and 2000 (Venerus et al., Reference Venerus, Galván, Irigoyen and Gosztonyi2007 and references therein).

This is the kind of analysis each case of unusual first sighting must go through before it can be considered a reliable support to other observations of climate changes.

Moving from Argentina across the Pacific, to the north-western Pacific, there is at least one recent case which is likely to be caused by climate change. This is an interesting case because it is a report of change in habitat use by a coral reef fish, but not a change in the geographical distribution pattern of this species (as most examples in this review are). In the Ryukyus Archipelago, southern Japan, at the southernmost local eastern reef of Akajima Island (26.640′N 127.867′E) Arvedlund & Takemura (Reference Arvedlund and Takemura2005) observed one adult anemonefish (SL = 9 cm) of the species Amphiprion clarkii (sensu Fautin & Allen, Reference Fautin and Allen1997) residing in the same soft coral of the genus Lobophytum sp. (sensu Fabricius & Alderslade, Reference Fabricius and Alderslade2001) at 1-m depth, over 20 months from May 2003 to December 2004 (Figure 1). Anemonefish comprise a well known guild of 28 damselfish (family Pomacentridae) that are obligate symbionts of 10 host sea anemones (families Actiniidae, Stichodactylidae and Thalassianthidae) in the tropical and suitable subtropical Indo-Pacific region (Allen, Reference Allen1975; Dunn, Reference Dunn1981; Fautin & Allen, Reference Fautin and Allen1997). This case can be traced back to a result of climate change. The 1998 global coral bleaching event (Wilkinson, Reference Wilkinson1998) had a serious impact on the local coral reefs of Sesoko Island (Loya et al., Reference Loya, Sakai, Yamazato, Nakano, Sambali and van Woesik2001). In addition, several host sea anemone species disappeared and the abundance of the surviving species declined seriously (Hattori, Reference Hattori2002). In February 2005 the host sea anemones had not recovered yet (Arvedlund, personal observation).

Fig. 1. First records of marine species of fish in highly unusual habitats are on the increase. The image shows the first long term record of the anemonefish Amphiprion clarkii in association with a soft coral of the genus Lobophytum at a shallow coral reef in the Ryukyu Archipelago, Japan, during 20 months in 2003 and 2004 (Arvedlund & Takemura, Reference Arvedlund and Takemura2005). This unusual anemonefish microhabitat pattern is thought to be the result of all host sea anemones dying out on local reefs. The 1998 global coral bleaching event (Wilkinson, Reference Wilkinson1998) had a serious impact on Japanese coral reefs (Loya et al., Reference Loya, Sakai, Yamazato, Nakano, Sambali and van Woesik2001). Image by M. Arvedlund.

Of cases which are most likely not indicators of climate changes, there is for example the anterior portion of a smooth hammerhead shark (Sphyrna zygaena), a large marine top predator, that was found near the high tide mark in Portreath Harbour, north Cornwall, the United Kingdom, by Mr Kevin Melton of Portreath (Southall & Sims, Reference Southall and Sims2006). This large hammerhead shark species has a broad geographical distribution, occupying coastal–pelagic and semi-oceanic habitats of the continental and insular shelves in tropical and warm-temperate zones of the Pacific, Atlantic and Indian Oceans. In the eastern North Atlantic it is present in the Mediterranean, extending south to Senegal, and northwards to the British Isles. It is considered exceptionally rare in British waters. Only five records of single individuals occurring in British waters exist, made between 1829 and 1865 (Southall & Sims, Reference Southall and Sims2006). However, because of these five records in the 19th Century, the recent record shows that the range pattern for this shark sometimes includes the United Kingdom, not because of climate change but for other reasons.

Two individuals of the longfin mako shark (Isurus paucus), a large marine top predator, were caught in September 2005, north-west of the Azores (42°50′N 36°16′W) on swordfish long lines, with baits set at a depth between 50 and 100 m over an average water depth of 4000 m (Queiroz et al., Reference Queiroz, Araújo, Ribeiro, Tarroso, Xavier and Santos2006). The longfin mako is a pelagic, oceanic, probably circumtropical shark, apparently common in the western Atlantic (Queiroz et al., Reference Queiroz, Araújo, Ribeiro, Tarroso, Xavier and Santos2006). Querioz et al.'s result confirms the presence of I. paucus in mid-North Atlantic waters, expanding their known distribution. The ragged-tooth shark (Odontaspis ferox) was recently spotted at Fernando de Noronha Archipelago, western equatorial Atlantic for the first time (Garla & Júnior, Reference Garla and Júnior2006). The ragged-tooth shark is a poorly known species with cosmopolitan and disjunct distribution throughout most of the tropical and subtropical oceans of the world (Garla & Júnior, Reference Garla and Júnior2006). These shark cases are most likely not indicators of climate changes because these fish are pelagic, and often cosmopolitan in range.

Another type of first records, which is of little or no use as an indicator of climate changes, is when marine fish with a prolonged pelagic larval phase are involved in first records of unusual distributions. Palacios-Salgado et al. (Reference Palacios-Salgado, Ramírez-Valdez and Rojas-Herrera2007) published a first record of Acanthurus nigricans (Perciformes: Acanthuridae) and Cantherhines dumerilii (Tetraodontiformes: Monacanthidae) from the Guerrero coast, México. Both species are categorized as wandering residents of the tropical eastern Pacific (Robertson et al., Reference Robertson, Grove and McCosker2004). Species of the genus Acanthurus are characterized by a prolonged larval stage, lasting several months (McCormick, Reference McCormick1999). The duration of the pelagic larval stage of species of the genus Cantherhines is unknown, but juveniles of Cantherhines dumerilii are often associated with floating objects (Castro et al., Reference Castro, Santiago and Santana-Ortega2002), which enables them to disperse widely (Palacios-Salgado et al., Reference Palacios-Salgado, Ramírez-Valdez and Rojas-Herrera2007 and references therein).

Finally, some species cannot be used as indicators because they have been transported directly by human activity into a new area and/or habitat. For example, the first record of the muzzled blenny, Omobranchus punctatus, was recently recorded for the first time in the Mediterranean Sea, collected among fouling organisms covering the frame of aquaculture cages at the entrance to the port of Ashdod, Israel (Golani, Reference Golani2004). Golani suggests that its presence there is most likely due to accidental introduction attributed to ship-mediated transfer. Such cases are numerous.

Summing up, some first records may be useful for marine ecologists, and fisheries and conservation management, in order to improve our understanding of declining and migrating commercially important populations of fish, provided each record is critically assessed and combined with recent records of fish population parameters, population dynamics, marine habitat assessments and abiotic data such as sea bottom, middle and surface temperatures.

More research and monitoring of climate change impacts on marine fish species and their populations are essential. The distribution and occurrence of populations should be surveyed and the migratory and reproductive behaviour of individuals has to be observed and recorded. I suggest that the establishment of easily accessible Internet online databases, containing first records of species of fish and GIS maps indicating where the first records were made, should be considered in the near future in order to help marine ecologist fisheries and conservation management track recent changes in fish populations quickly. Internet discussion fora covering first records and related issues may also be of help. DNA barcodes on as many species as possible are also a good technique to show whether fish populations are closed or open, i.e. the connectivity of populations. This may help to show if a first record of a fish species in an unusual habitat or a new geographical area is an indicator of climate changes or just a natural part of this species dispersal pattern. For example, tropical demersal coral reef fish populations were until recently considered open. The pelagic nature of the early larval stages of marine fish should point to a passive dispersal model where ocean currents cause panmixis over great spatial scales and prevent isolation of populations, a precondition for speciation (Gerlach et al., Reference Gerlach, Atema, Kingsford, Black and Miller-Sims2007). Even though ocean currents are a major force in larval dispersal, recent studies show far greater retention than predicted by advection models (Gerlach et al., Reference Gerlach, Atema, Kingsford, Black and Miller-Sims2007).