Introduction

Subtidal canopy-forming kelps and fucoids form impressive underwater forests across temperate and polar latitudes. Declines and local extirpations of these seaweeds are however becoming increasingly prevalent as anthropogenic impacts increase on the marine environment (Thibaut et al. 2005; Krumhansl et al. 2016; McPherson et al. 2021), with negative consequences for ecosystem functioning and their associated fauna and flora. Losses and local extirpations have been particularly widespread in areas with intense human pressure such as heavily urbanized coasts (Kautsky et al. 1992; Thibaut et al. 2005), with non-urbanized areas typically avoiding most declines (Benedetti-Cecchi et al. 2001; Scherner et al. 2013). Yet, studies on subtidal canopy-forming algae have mostly focused on charismatic and well-studied species (mostly kelps), with little being known about other canopy-forming groups (Shepherd and Edgar 2013). Forests formed by fucoid algae (order Fucales) are more globally widespread and diverse than kelp forests (Fragkopoulou et al. 2022), but remain comparatively understudied (Coleman and Wernberg 2017). Importantly, fucoid forests play important roles in supporting temperate biodiversity (Taylor and Cole 1994; Fraser et al. 2020), as well as nearshore food webs and coastal carbon cycling (Gouvêa et al. 2020; Pessarrodona and Grimaldi 2022).

Temperate Australia features one of the richest marine floras in the world, including an unparallel and highly endemic diversity of forest-forming algae (> 60 species). A range of anthropogenic pressures has precipitated declines and local extirpations of several of these species (Coleman et al. 2008; Connell et al. 2008; Shepherd et al. 2009; Phillips and Blackshaw 2011; Smale and Wernberg 2013; Wernberg et al. 2016), with ocean warming threatening severe range contractions in the coming decades (Martínez et al. 2018). Of particular concern is the predicted contractions of species in the genus Cystophora, which is the second largest genus of fucoids globally (23 species, after Sargassum; Guiry and Guiry 2022) and is endemic to Australasia.

Cystophora spp. constitute a widespread component of the temperate rocky subtidal of Australasia, occurring from the intertidal to around 30 m depth (Bennett 1921; Womersley 1947, 1949; Smith 1952). Most species are monoecious and have a marked seasonal life cycle (Hotchkiss 1999), growing at different times than other canopy-forming groups (“season anticipator” strategy, Pessarrodona and Grimaldi, 2022). Plants are perennial, reaching the highest biomass in late winter/early spring when egg maturation and release occur (Klemm 1988; Hotchkiss 1999), with upwards of 75% of the accumulated biomass being lost by summer (Pessarrodona and Grimaldi, 2022). Their standing biomass and productivity can rival than that of the more well-studied kelp forests, although their stand structure and seasonal productivity are distinct suggesting they play a functionally unique role from that of other canopy-forming seaweeds (Pessarrodona and Grimaldi, 2022). Cystophora spp. tend to dominate moderately-exposed and sheltered locations (Shepherd and Womersley 1981; Collings and Cheshire 1998; Turner and Cheshire 2003; Goldberg and Kendrick 2004; Turner 2004; Pessarrodona and Grimaldi 2022), with plants often occurring in mixed stands with no single species dominating the assemblage (Turner and Cheshire 2003), much like trees in tropical rainforests. Despite being the prevalent canopy-forming algae along much of the southern Australian continent (Turner and Cheshire 2003; Goldberg and Kendrick 2004), little is known about their ecology and fine-scale biogeographic distribution. Here, I combine presence/absence records from herbaria and field surveys to show that the entire genus has largely disappeared from the warmer range of its distribution in Western Australia (Fig. 1).

Fig. 1
figure 1

Occurrence of Cystophora spp. either as drift or on the reef across the western range edge of its distribution (depicted in green) over time. Presences are indicated in filled black dots, while unfilled dots denote absences in surveyed locations. The extent of the putative functional range contractions of several Cystophora spp. is shown by a red arrow. Note that previous to the advent of subtidal work in the area in the 1980s, records relied on drift specimens and are mostly from herbaria collections and therefore only presences are reported

Full size image

Methods

To establish the historical biogeographical distribution of Cystophora across its western range edge, I conducted a comprehensive search of Cystophora records in herbaria, government reports, and peer-reviewed studies. Historical herbaria collections were sourced from the Australasian Virtual Herbarium (AVH 2021) and the Macroalgal Herbarium Consortium Portal (macroalgae.org), which contains digitized information of herbaria stored in the USA. I also searched for records in the Natural History Museum (UK) and the Muséum National d’Historie Naturelle (France) given that many French and British botanists collected macroalgal specimens from Australian coasts (Womersley 1959). However, specimens in those collections were not properly georeferenced and therefore not included. More recent presence/absence records were gathered from the grey literature (Gordon 1986; Moore 1987; Walker et al. 1991, 1992; Burt and Anderton 1997; Colman 1997; Bancroft and Davidson 1998; Kendrick et al. 1999; Westera et al. 2007; Edgar et al. 2009), published datasets (CSIRO 2005), peer-reviewed studies (Kendrick et al. 2004; Waddington et al. 2010; Richards et al. 2016), and unpublished theses (Smith 1952; Wood 1980; Phillips 1996; Scott 2012).

Finally, a series of 30 min underwater visual censuses were conducted between 2018 and 2022 to determine the contemporary extent of Cystophora species (79 survey sites from Geraldton to Esperance). These surveys specifically targeted Cystophora along the entirety of the coast—including revisits of many of the historical collection sites (Table 1)—to complement observations from ecologically-focused studies by other authors that may have missed them or worked at a coarser-taxonomic level. The majority of surveys (65%) targeted the upper subtidal (0.1 m) up to 3 m depth, as that is their reported habitat throughout their historical and current range (Smith, 1952, PERTH5857759, AD-A51087, AD-A33330B, HO594202). Additionally, to check whether Cystophora spp. may have contemporarily retreated to greater depths, I conducted 27 surveys on SCUBA targeting a deeper (6–12 m) depth range, including historical sites but also new ones (Table 1). Finally, as some of the earliest collections were from plants washed ashore on the beach, I also conducted macroalgae wrack surveys after heavy storms. On these, the author walked for 1 km along the wrack line examining all the canopy-forming algae encountered along transect (n = 6 sites; Table 1).

Table 1 Details of the 79 Underwater Visual Surveys conducted specifically targeting Cystophora spp. Note that some localities did not have historical records (non-available, NA), but were in-between locations that did and were also surveyed. The number of sites surveyed within each depth range is indicated in parenthesis
Full size table

Results and discussion

According to historical reports and specimens deposited in herbaria (Lucas 1936; Womersley 1964, 1987; AVH 2021), a total of sixteen Cystophora species occur in Western Australia. Of these, six species were not considered as they had ≤ 10 available historical records (C. siliquosa, C. gracilis, C. botryocystis, C. polycystidea, C. tenuis) or extremely narrow ranges (C. harveyi, ≈ 300 km). Before the advent of SCUBA in the study area in the 1980, records relied on collections of shallow individuals or drift material. During the 1950s–1970s, Cystophora plants were frequently collected (> 100 records) as drift by phycologists at a variety of locations across the warm range edge (e.g., Dongara, 29°S 115°E; Yanchep, 31°S 115°E; Fremantle, 32°S 115°E), as well as throughout the colder South coast (> 300 records overall) (Hodgkin 1959a, b; Womersley 1964; AVH 2021). The genus range edge was located as far north as Port Denison/Geraldton (29°S 115°E; C. brownii, C. grevillei, and C. monilifera), while several other species extended up to the Perth Metropolitan area (32°S 115°E, e.g., C. pectinata, C. subfarcinata, C. retroflexa; Fig. 1). The pioneering studies around that area by G. G. Smith (1952) described five species (C. brownii, C. grevillei, C. monilifera, C. pectinata, C. retroflexa) being present as “typical subordinate species of the [shallow subtidal] reef system” during the 1940–1950s, with C. monilifera and C. brownii also extending to the intertidal-subtidal fringe.

The contemporary (2000–2022) distribution of the genus revealed minimal distributional changes occurred for species inhabiting cooler portions or the genus range (i.e., 20 °C annual mean surface temperature, ca. below 32.5°S latitude), while seven out of ten species appeared to be absent from the historical range edge (Fig. 1). Of these, five species (C. brownii, C. grevillei, C. pectinata, C. monilifera, and C. subfarcinata) had multiple records in that area (n > 3) and contained specimens collected or identified by the eminent Australian phycologist H.B.S Womersley. They were therefore considered to be reliable historical records and examined further.

By the time some of the first subtidal work on SCUBA started in the 1980s, the number of overall records of these species in the warm range edge had drastically declined (4 records across 175 sampling events at 175 sites, Fig. 2). Intense sampling (40 sites) of the algal communities in the Marmion lagoon (32°S 115°E) for the establishment of the state’s first marine park failed to detect any species of Cystophora (Moore 1987), and so did studies conducted at Perth’s nearshore reefs similar to the ones where they were historically frequent (Phillips 1996). The early 2000s saw a sharp increase in the algal survey efforts which translated in an increase in Cystophora records across the entire coast (> 300 records across 300 sampling events at 228 sites; Fig. 2), but not in the warm range edge (7 records across 191 sampling events and 147 sites). A few sparse individuals of C. brownii (0.8–1.7% cover) were recorded at two of the 42 sites surveyed between Jurien Bay and Lancelin (30–31°S 115°E; Barrett et al. 2002; Edgar et al. 2005), and single specimens of C. grevillei and C. monilifera were recorded in Rottnest Island (32°S 115°E; Scott 2012; AVH 2021). It thus appears that Cystophora was rare or had disappeared from the most of its warm range edge by the early 2000s.

Fig. 2
figure 2

Number of records of Cystophora spp. over time along the western rear (warm) edge (> 20 °C annual mean surface temperature, ca. above 32.5° S latitude) and central (cool) range (< 20 °C, below 32.5° S latitude). Records include herbaria collections, field algal surveys from the peer-reviewed and grey literature, as well as underwater visual censuses (UVCs) targeting Cystophora across the entirety of its range (2018–2022). The high number of records during the 2000–2022 in the south is largely due to the UVCs conducted in this study and biodiversity surveys across seasons (Westera et al. 2007; Edgar et al. 2009)

Full size image

The 2010s were marked by a prolonged marine heatwave that negatively affected Western Australian marine ecosystems, driving the range contraction of several cold-water seaweeds (Smale and Wernberg 2013; Wernberg et al. 2016). A comprehensive set of algal surveys along 52 reefs conducted immediately after the heatwave only recorded Cystophora at the cooler (i.e., < 20 °C annual mean surface temperature) portions of its distribution (Bennett 2015). During the Cystophora visual censuses I conducted a decade after the heatwave (2018–2022), Cystophora was absent at eight of the nine historical sites where it was present on the reef in the warm range edge (Fig. 3a), including the well-studied sites around the Perth Metropolitan area (Fig. 3b). In contrast, all abundant Cystophora species occurring in Western Australia were recorded from the cooler distributional range, and Cystophora was present in all the twelve historical sites (Fig. 3a). Although it remains possible that small, isolated populations exist in warm areas that were not sampled throughout all these years, as suggested by the few isolated C. brownii plants that the author recorded in three of the 43 sites examined in the warm range edge (Cervantes, 30.6°S 115.1°E; Seabird, 31.3°S 115.4°E; Natural Jetty Reef, 32.0°S 115.5°E), it appears that Cystophora species may thus be functionally extinct across ca. 100–300 km of their historical range edge. These losses far exceed those documented for kelps (Wernberg et al. 2016) and other fucoids (Smale and Wernberg 2013) in the study area (Fig. 4). In the case of C. pectinata, this contraction equates to a loss of approximately 8% of its global distribution. Understanding how the decline of Cystophora spp. has affected the ecology of shallow subtidal reefs is thus a priority for future research.

Fig. 3
figure 3

Results of resurveys where Cystophora spp. were historically documented to be present on the reef (circle) or as drift (triangles). Filled symbols indicate presences on the resurveys, while absences are indicated by unfilled symbols. Note that additional surveys at other sites were conducted but these are not shown as they lacked historical records

Full size image
Fig. 4
figure 4

Apparent range contractions (km of coastline) of fucoid (brown) and kelp (yellow) forest-forming species in the study area documented in this study and the literature. Losses were determined by measuring the coastal distance between historical and contemporary populations of each species

Full size image

The approach used here necessitated collection of records from several sources, which may have varied in sampling effort. Herbarium records rarely represent systematic collecting efforts, and so it is not possible to estimate the historical Cystophora abundance throughout their warmer range nor determine whether they had a continuous distribution. The frequency in which they were encountered along the coast however suggests that they were common, at least until the past half-century (Smith 1952; Hodgkin 1959a). Another challenge with historical records is that they often rely on drift specimens, and so it is possible that they represent presences of individuals transported from elsewhere, giving misleading impressions of the actual historical range edge. This is unlikely the case here however as most of the species whose range presumably contracted had either been documented on the reef (C. grevillei, C. pectinata) or do not have floating structures that would seemingly allow for long-range transport (C. brownii, C. retorta, C. subfarcinata). An additional issue with collating presence/absence data from surveys with other objectives is that untrained scientists may have failed to identify Cystophora (i.e., false absences), yielding misleading impressions of their contemporary decline. This is, however, highly unlikely. Many of the contemporary surveys conducted during that time exclusively targeted algal assemblages (Phillips 1996; Wernberg et al. 2003; Westera et al. 2007; Smale et al. 2010; Bennett 2015), some of which recorded upwards of 250 species of macroalgae. The recent sampling effort across the warm range edge also increased massively in terms of spatial cover and number of sampling events (> 400 sampling events at > 300 sites after the 1980s versus 76 sampling events at 71 sites before). Additionally, surveys conducted at the same area on multiple occasions reported Cystophora repeatedly, even when algae was not the sole focus on the survey (Edgar et al. 2009). Finally, the visual censuses that I conducted specifically targeted Cystophora and included some of the historical collection sites, reinforcing evidence that the genus has declined substantially.

While declines and local extirpations of canopy-forming macroalgae have been recorded worldwide (e.g., Coleman et al. 2008; Connell et al. 2008; Phillips and Blackshaw, 2011; Thibaut et al., 2015), the apparent range retractions documented here are particularly unique in that they involve the functional extinction of multiple species within the same genus across large sections of coastline. Analysis of herbaria historical records shows that numerous seaweeds in the study area may have also experienced declines in recent decades (Wernberg et al. 2011), which suggests their decline adds to local-scale losses in diversity. The niche left by the disappearance of Cystophora spp. may have been filled by other canopy-forming seaweeds such as the kelp Ecklonia radiata, which typically competes for space with Cystophora across South Australian subtidal reefs (Turner 2004). Another possibility includes their replacement by warmer-affinity canopy formers (e.g., some species of Sargassum); this seems likely as these two genera often coexist (Collings 1996; Goldberg and Kendrick 2004; Turner 2004), albeit further studies on the distribution and taxonomy of Sargassum in the study area are needed to validate this possibility. In any case, given that Cystophora spp. have a stand structure, biomass, and growth pattern that differ from that of Ecklonia or Sargassum (Irving and Connell 2006; Pessarrodona and Grimaldi 2022), it is likely that their demise has adverse impacts for ecosystem functioning.

While it was not possible to determine the exact causes of Cystophora spp. demise in this study, there are several non-mutually exclusive possibilities. Summer temperatures are the best predictor of Cystophora spp. biogeographical distributions (Martínez et al. 2018), suggesting that they are highly sensitive to warming. Indeed, anomalously high water temperatures have been posited to be responsible for Cystophora recruitment failure in South Australian populations (Turner 2004). Along the warmer West Australian coast, Cystophora biomass declines sharply along a 3 °C latitudinal gradient, hinting its abundance is strongly linked to cool-water temperature (Bennett 2015). Since the 1950s, the waters around the Western Australian continental shelf have warmed about 0.9 °C (Pearce and Feng 2007) and experienced an extreme marine heatwave in 2011 that precipitated multiple seaweed range contractions (Smale and Wernberg 2013; Wernberg et al. 2016). It thus seems plausible that warming is one of the lead causes of their decline, although causal experimental links between temperature and Cystophora performance remain to be established. Another possible driver of decline is decreasing water quality associated with urbanization. While the majority of the Western Australian coast remains relatively unurbanized, urban development since the 1950s has concentrated in the Perth metropolitan area, whose population has grown seven-fold. Multiple ocean outlets have been added since the 1960s to satisfy the increasing wastewater supply, which discharged ca. 80 gigaliters year−1 around the 2000s. While the tolerance of Cystophora to increased nutrients is unknown, other Australian subtidal fucoids are highly sensitive to treated sewage effluents (Brown et al. 1990). An additional driver could be increases in sedimentation, which have driven declines in some Cystophora spp. along sheltered areas and embayments (Shepherd et al. 2009). Sediment stress is unlikely to be the main driver in Western Australia however, as reefs in the region are exposed to intense wave action and large oceanic swells (Hemer 2006), which generally maintain low levels of turbidity and sedimentation. Additionally, Cystophora spp. appear to be somewhat more tolerant to pulse sediment disturbances (e.g., sediment plumes) than other types of canopy-forming seaweeds (Turner 2004).

Recovery of Cystophora spp. across their former range seems unlikely given that only reduced populations of few individuals like the one encountered in Cervantes or Seabird possibly remain, severely restricting the available propagule supply. The large distances between remaining populations may further prevent recolonization and connectivity. Over 90% of the propagules from fucoids settle within meters of the parental thalli (Kendrick and Walker 1995), Cystophora recruitment being negligible 500 m away from reproductive plants (Goldberg et al. 2004). The functional extinction of Cystophora along its warm range edge is concerning, as they play functionally unique roles in the ecology of temperate reefs (Coleman and Wernberg 2017), supporting different floral and faunal assemblages (Taylor and Cole 1994; Irving and Connell 2006; McDermott and Shima 2006; Fraser et al. 2020), and having different growth patterns than the more well-studied kelp species for example (Pessarrodona and Grimaldi 2022). Further warming and coastal development threaten Cystophora populations established along the cooler southern coast (Martínez et al. 2018), which is concerning given that several species are endemic to that area (e.g., C. gracilis, C. harveyi) and have extremely narrow geographical distributions (e.g., 300 km for C. harveyi). A better understanding of their ecology is key to predict their response to increasing anthropogenic pressures and prevent further declines.