Quaternary Research (2026), 130, 59–69 doi:10.1017/qua.2025.10051 Research Article Spatial heterogeneity in Holocene vegetation dynamics across Bass Strait and its regional paleoclimatic implications Matthew Adesanya Adeleyea , Simon Graeme Haberleb,c,d, Grant Williamsone and David Bowmane aDepartment of Geography, University of Cambridge, Cambridgeshire, CB2 3EL, UK; bSchool of Culture, History and Language, College of Asia and the Pacific, The Australian National University, Canberra, ACT 2601, Australia; cAustralian Research Council Centre of Excellence for Australian Biodiversity and Heritage, College of Asia and the Pacific, The Australian National University, Canberra, ACT 2601, Australia; dAustralian Research Council Centre of Excellence for Indigenous and Environmental Histories and Futures, College of Asia and the Pacific, The Australian National University, Canberra, ACT 2601, Australia and eFire Centre, School of Natural Sciences, University of Tasmania, Sandy Bay, TAS 7001, Australia Abstract We present a Late Pleistocene paleoecological record from King Island in western Bass Strait, Tasmania, and compare this to existing records from the eastern Bass Strait islands to improve our understanding of the region’s paleoecology and paleoclimatology. Vegetation change across the region followed similar trajectories during the late glacial–Middle Holocene, characterized by homogeneous warming and wetting trends. Spatial divergence occurred during the Middle Holocene when sea level rose, and different drivers began influencing western and eastern Bass Strait islands. In eastern Bass Strait, Middle Holocene sea-level rise caused replacement of woodland by coastal heathland, while in the west, a drier period accompanied by fires transformed forests to forest–scrub. The comparative analysis suggests that Westerly driven climatic anti-phasing was pronounced at higher latitudes of Tasmania during the late glacial–Early Holocene. A combination of weak Leeuwin Current, positive Indian Ocean Dipole (IOD), and El Niño–Southern Oscillation (ENSO) intensification contributed to Middle Holocene aridity across Bass Strait. Strong Westerlies and negative IOD phases led to greater regionalization of rainfall across western Bass Strait during the Late Holocene, while ENSO intensification drove rainfall declines in eastern Bass Strait. These findings provide new insights into the complexity of Late Pleistocene environmental dynamics across southeast Australia. Keywords: Australia; Bass Strait; Holocene; Indian Ocean Dipole; King Island; Leeuwin Current; paleoecology; paleoclimate; Tasmania Introduction The Holocene climate history of southeast Australia is not well understood, especially at the sub-regional level where dynamic coastlines, arid inland plains, and stable mountain ranges may have driven local variability in hydroclimate patterns through time. Southeastern Australia includes the now submerged Bassian Plain (currently Bass Strait) that formed a land bridge between Victoria and eastern South Australia and Tasmania (Aboriginal- given name: Lutruwita) during much of the last glacial period (Lambeck and Chappell, 2001). The west–east distribution of islands in the Bass Strait makes them ideal locations to investigate the history of the general wet west to dry east climatic gradi- ent that currently characterizes this part of southeast Australia. Previous studies from the area suggest a synchronous dry Middle Holocene on the western and eastern islands (Adeleye et al., 2021a, 2025), which contrasts with previous hypotheses of a more regional wet climate (De Deckker, 2022). The spatio-temporal response of terrestrial vegetation communities to past climate between the climatically different islands of western and eastern Corresponding author: Matthew Adesanya Adeleye; Email: ma2073@cam.ac.uk Cite this article: Adeleye, M.A., Haberle, S.G., Williamson, G., Bowman, D., 2026. Spatial heterogeneity in Holocene vegetation dynamics across Bass Strait and its regional paleoclimatic implications. Quaternary Research 130, 59–69. https://doi.org/10.1017/ qua.2025.10051 Bass Strait is poorly known. This knowledge is key to under- standing the paleoclimate of the Bass Strait region, and how these records fit within the broader paleoclimatic patterns for southeast Australia. Studies of Holocene wetland development on Cape Barren Island (Aboriginal-given name: Truwana) in eastern Bass Strait and King Island in western Bass Strait suggest a pattern of Middle Holocene dryness based on a shift from permanent deeper water wetland taxa towards more shallow-water-tolerant wetland taxa (Adeleye et al., 2021a, b). This dry episode coincides with the timing of regional sea-level transgressions in southeast Australia (Sloss et al., 2007; Dougherty et al., 2019), which also drove expan- sion of coastal vegetation communities on Cape Barren Island (Adeleye et al., 2021b) and nearby areas (Thomas and Kirkpatrick, 1996; McWethy et al., 2017). The onset of more frequent El Niño–Southern Oscillation (ENSO) episodes during the Middle Holocene, however, is thought to have contributed to an overall dryness at this time (Adeleye et al., 2021a). Elevated fire activity in the Middle Holocene also occurred but was not as important as the effect of sea-level change on terrestrial vegetation communities (McWethy et al., 2017; Adeleye et al., 2021b). The effect that these dynamic regional paleoclimate changes had on the vegetation and wetland environments of King Island during the Middle Holocene (Adeleye et al., 2025) is less well understood. © The Author(s), 2025. Published by Cambridge University Press on behalf of Quaternary Research Center. This is an Open Access article, distributed under the terms of the Creative Commons Attribution licence (http://creativecommons.org/licenses/by/4.0), which permits unrestricted re-use, distribution and reproduction, provided the original article is properly cited. https://orcid.org/0000-0002-6034-5807 https://orcid.org/0000-0001-8075-124X mailto:ma2073@cam.ac.uk https://doi.org/10.1017/qua.2025.10051 https://doi.org/10.1017/qua.2025.10051 http://creativecommons.org/licenses/by/4.0 60 M.A. Adeleye et al. Further empirical knowledge of the temporal dynamics of ter- restrial vegetation communities on western and eastern Bass Strait islands will help us to better illuminate past sub-regional and regional climate of the areas, ultimately improving our under- standing of the Bass Strait landscape and climate of southeast Australia during the Holocene. Therefore, our main goal in this study is to contrast a new terrestrial paleoecological record from King Island with an existing comparable record on Cape Barren Island (Adeleye et al., 2021a, b) to investigate the spatio-temporal response of vegetation to Holocene climate variability in the wet west and dry east climatic regions of Bass Strait. Other climate drivers, such as sea-level change, and potential anthropogenic land-use effect also are considered. It is important to note in this context that there are few archaeological studies of past human occupation in the Bass Strait area. Of the two islands of focus here, existing evidence suggests more continuous occupation on Cape Barren Island in the east compared to King Island in the west during theHolocene (Sim, 1998; Bowdler, 2015).We therefore also consider the influence of human agency in shaping vegetation patterns throughout the Holocene. Methods Western Bass Strait: King Island study site At 1098 km2, King Island is the second largest of the Bass Strait islands after Flinders Island, which is 1367 km2. King Island sits in the western part of Bass Strait, northwest of Tasmania and south- east of Victoria, mainland Australia (Fig. 1). It receives the highest rainfall of all the Bass Strait islands yet shares a temperatemaritime climate with them. Year-to-year rainfall variability on King Island is largely correlatedwithwesterly circulation strength over the area, most strongly relating to the Southern Annular Mode (SAM), par- ticularly in summermonths (Fig. 1). In some seasons, the influence of El Niño–Southern Oscillation (ENSO) and the Indian Ocean Dipole (IOD) (Australian Bureau of Meteorology, undated) on monthly rainfall is apparent at the multi-decadal scale (Fig. 1). Early European accounts of King Island’s landscape indicate the island vegetation was dominated by communities of dense Leptospermum scrub and wet Eucalyptus forests. Since European settlement from the early 1800s, about 70% of the island’s native vegetation has been cleared for dairy- and beef-cattle produc- tion (King Island Natural Resource Management Group, 2002). The remaining native vegetation on King Island today comprises diverse communities of forest, scrub, heathland, and sedgeland. Large areas of scrub persist on the island and are dominated by Leptospermum and Melaleuca species (King Island Natural Resource Management Group, 2002). Heathland and sedgeland occupy the northern part of the island in poorly drained soils. Forest communities are now restricted to discontinuous patches of uncleared lands and are dominated by one or more of the follow- ing: Eucalyptus sp., Melaleuca ericifolia, and Acacia melanoxylon. Native vegetation composition on King Island is generally simi- lar to other western Bass Strait Islands. By contrast, eastern Bass Strait islands are dominated by dry scrub and forest communities. Also, unlike the wetter King Island where Melaleuca mostly form forests, on the drier eastern Bass Strait Islands species in this genus mostly grow as part of heathland and scrub communities, espe- cially so for Cape Barren Island, which currently hosts the largest heathland and scrub area in the region (Adeleye et al., 2021b). There is now clear evidence for a long history of human interaction with the Bass Strait islands and the now submerged glacial-period Bassian Plain spanning at least the last 41,600 years (Adeleye et al., 2024), although the nature and intensity of land use likely shifted over time in response to sea-level rise and climatic variability. Archaeological evidence directly from King Island shows human presence on the Bassian Plain during the glacial period, including a human burial site dated to approxi- mately 15,000 years ago (Sim and Thorne, 1990). Formation of the contemporary island coastline, with an open water distance to the nearest large island of Tasmania of around 65 km for King Island and 20 km for the eastern Bass Strait islands, appears to have played a part in reducing the likelihood of continuous occu- pation of these islands.Material culture preserved in shell middens and open archaeological contexts suggest continued use of the Bass Strait islands was maintained during the Early and Middle Holocene, as sea levels rose and stabilized, with evidence of occu- pation until at least ca. 12,000 years ago on King Island and as late as 4500 years ago in eastern Bass Strait islands (Sim, 1990; Bowdler, 2015). After this time, more intermittent use was likely during the Late Holocene, driven by connections to, and seasonal reliance on, island–maritime resources (Bowdler, 2015; Adeleye et al., 2024). At the time of the arrival of Europeans around the turn of the nine- teenth century, both King Island andCape Barren Island in the east were reported as being uninhabited (Sim, 1990; Sim and Thorne, 1990; Bowdler, 2015). Sampling A 3-m sediment core sequence was collected from Lake Martha Lavinia (39°38’52.0”S, 144°03’53.3”E; 20 m asl) on King Island using a D-section corer. Six bulk sediment samples (2.5 cm3) were analyzed for radiocarbon dating at DirectAMS, Bothell, Washington, USA. Resulting AMS radiocarbon dates were used to build a Bayesian age–depth model in ‘rbacon’ (Blaauw et al., 2022), with the latest calibration curve (SHCal20; Hogg et al., 2020). The lake was accessed by foot, and a core sequence was taken from two overlapping holes next to each other in the shallow part (∼0.8 m depth), about halfway to the lake center (∼2.0 m deep). We were unable to access the center due to strong wind speeds on the day of coring and the extremely soft lake floor, which made it inac- cessible by foot. Sediment samples (1.25 cm3) were taken at 2- to 4-cm intervals in the core for pollen analysis and 1-cm intervals for macrofossil analysis (primarily macrocharcoal) to reconstruct terrestrial vegetation and fire history. Preparation of pollen sam- ples included HCl, KOH, and acetolysis treatment (Faegri and Iversen, 1975), while sediment subsamples were soaked in house- hold bleach for ca. 15 hours and washed through a 125-µm sieve for macrofossil analysis. At least 300 terrestrial pollen grains were identified and counted in each sample. Data analysis Squared Chord Distance (SCD) dissimilarity between pollen sam- ples was calculated using the ‘analogue’ package in R (Simpson et al., 2024). This was to estimate compositional turnover of taxa in pollen records, which indicates the magnitude of change in vegeta- tion over time (Davis et al., 2015).The empirically derived turnover threshold for southeast Australia (SCD = 0.2) based on modern pollen–vegetation relationship across the region (Adeleye et al., 2021c) was used to determine the timing of major turnover and vegetation community phases in the pollen record. An SCD value > 0.2 means a complete shift in pollen assemblages and an SCD< 0.2 means no significant shift in pollen assemblages. Holocene vegetation dynamics, Bass Strait, Tasmania 61 Lake Martha Lavinia (b) King Island King Isl. Flinders Isl. Cape Barren Isl. (Truwana) (a) Bass Strait Tasmania (Lutruwita) AUSTRALIAAUSTRALIA Rainfall (mm) 0 700 1400 2100 2800 >3500 In d ia n O ce a n P U O R G X U A E N R U F FLEURIEU GROUP (c) (d) (e) (f) (g) In dia n Oc ea n Indian Ocean 0 20 40 60 80 Miles Figure 1. Study area and coring site on King Island (a, b). Rainfall map (a) was generated from Land Information System Tasmania (Department of Natural Resources and Environment Tasmania, 2021). (c–e) Correlation of map of monthly multi-decadal rainfall with climatic modes over Tasmania (Lutruwita) from 1957–2021, including the Southern Oscillation Index (SOI): positive wetter, Indian Ocean Dipole (IOD): positive drier, and Southern Annular Mode (SAM): positive drier. (f, g) Correlation of SAM and rainfall during Austral winter (JJA) and summer months (DJF). The pollen record was also rarified using the ‘vegan’ package in R to estimate palynological richness (Oksanen et al., 2024), which indicates floristic richness (Birks et al., 2016). This method is known to have inherent limitations, such as differential pollen productivity (Meltsov et al., 2011), although studies in Australia (Adeleye et al., 2021c) and in the Northern Hemisphere (e.g., Meltsov et al., 2013; Connor et al., 2021) have shown pollen rich- ness to generally reflect vegetation floristic richness, especially at landscape scales. In order to assess the convergence and divergence in vegetation trajectories between the King Island site (Lake Martha Lavinia) and a previously studied site on Cape Barren Island (Big Reedy Lagoon), we first applied Detrended Correspondence Analysis (DCA) to square-root-transformed pollen data from each site and then interpolated results to a mean temporal resolution of 250 years across sites from 0–14 cal ka BP (all ka ages below are cali- brated). The resulting ordination scores were then compared using Procrustes analysis and a permutation-based test (PROTEST, 999 permutations) was used to assess the statistical significance of the concordance between the sites.These analyses were also conducted in R using the ‘vegan’ package (Oksanen et al., 2024). The charcoal influx record (charcoal pieces/cm2/yr) for Lake Martha Lavinia was already presented in Adeleye et al. (2025). However, here we included the fire (peak) frequency estimate for the site, determined using the CharAnalysis program (Higuera et al., 2009, 2010). Given the multi-millennial scale of study, fire 62 M.A. Adeleye et al. frequency here is the frequency of fire episodes and not individ- ual fire events, which was calculated per 500 years, with minimum count cut-off probability for peak detection set to 0.05 using a Gaussian mixed model. Results Lake Martha Lavinia, King Island The Lake Martha Lavinia age–depth model shows an estimated basal age of ca. 14.8 ka BP and mean sediment accumulation rates of 0.02 cm/yr, with the highest accumulation at ca. 7.2–3.5 cal ka and the most recent ca. 700 years (Table 1; Fig. 2). Turnover analy- sis and threshold (SCD = 0.2) identified four terrestrial vegetation community phases: LML-1 (ca. 15–13 cal ka BP), LML-2 (ca. 13–12 ka), LML-3 (ca. 12–5.5 cal ka), and LML-4 (ca. 5.5 ka to present) (Fig. 3a). Herbaceous communities dominate LML-1 (ca. 15–13 ka) at ∼60%, with Eucalyptus and Banksia making up just under ∼40%. LML-2 (13–12 ka) sees the transition from herbaceous- to woody-dominated vegetation (Fig. 3a). Eucalyptus and Melaleuca increased substantially in this phase, reaching a total of ∼90%, while Banksia declined. This generally continued into LML-3 (12–5.5 ka), but with an increase in Acacia (∼20%) and minor increases in plants such as Leptospermum, Monotoca, and ferns, especially Dicksonia antarctica. Eucalyptus initially dropped to ∼10% in LML-4 (ca. 5.5 ka to present) in favor of Leptospermum (∼30%) and Melaleuca (∼40%). Acacia decreased to < 10% as well and never recovered. Melaleuca and Leptospermum later fell to < 20% in favor of Eucalyptus (∼20%) and Monotoca (∼20%). Minor increases in other taxa, such as Allocasuarina, Pomaderris, and Dodonaea were also observed and floristic richness peaked in this phase, reaching over 30 taxa/minimum count (Fig. 3a). Charcoal influx was high in Lake Martha Lavinia before 12 ka and between ca. 7 and 5 ka. However, influx was greatest in the for- mer and so was fire frequency, reaching 150 particles/cm2/yr and >2 fire episodes per 500 years, respectively. This period also coin- cides with the timing of the greatest turnover in the pollen record, with SCD > 0.3 (Fig. 3a). Wetland plants Restionaceae and Cyperaceae characterized the margin of Lake Martha Lavinia between ca. 15 and 10 cal ka, and freshwater green algae Zygnemataceae were abundant in the water, which is indicative of high wetland water levels (Fig. 3b). The wet- land plants and algae declined after this time, falling to aminimum between 9 and 5.5 ka. Bog fungal type Microthyriaceae increased during this period, which is reflective of low wetland water levels. The previous wetland plant community (andwater level) recovered after 5.5 ka, with Triglochin plant and other wetland fungi (e.g., Helicoon) and invertebrates (e.g., Copepoda, Assulina) appearing in the wetland assemblage (Fig. 3b). Big Reedy Lagoon, Cape Barren Island The pollen record from Big Reedy Lagoon on Cape Barren Island, eastern Bass Strait, shows a Poaceae-dominated herba- ceous community in the area before 13 ka (Adeleye et al., 2021b). Eucalyptus and Melaleuca expanded from 13 ka, with increased charcoal accumulation from ca. 11.7 ka.Wetland plants Isoetes and Cyperaceae locally dominated Big Reedy Lagoon from ca. 12–9 ka (Adeleye et al., 2021a). Allocasuarina increased at the expense of Eucalyptus from ca. 9.5 ka. In addition to the background abundance of Melaleuca and Allocasuarina, Sprengelia and Epacris dominated during ca. 9–6 ka and charcoal accumulationdecreased. Restionaceae replaced Isoetes and Cyperaceae in the wetland dur- ing this period. Sprengelia and Epacris decline favored increases in Leptospermum, Monotoca, and Pomaderris after 6.5 ka. Cyperaceae recovered in the wetland while Restionaceae decreased, with Botryococcus algae later joining the wetland community in the Late Holocene (Adeleye et al., 2021a, b). Comparing Lake Martha Lavinia and Big Reedy Lagoon Detrended Correspondence Analysis (DCA) and Procrustes anal- yses show a significant similarity in the King Island (Lake Martha Lavinia) and Cape Barren Island (Big Reedy Lagoon) sites pollen assemblages (m12 squared = 0.58, correlation = 0.65, p-value = 0.001). This similarity is most evident in the low divergence between 14 and 6 ka. High divergence is observed between sites from 6 ka to present (Fig. 4). Discussion This palynological study advances the understanding of the cli- mate dynamics across the Late Pleistocene Bassian Plain and the Holocene Bass Strait, particularly ranking the relative strength of major interannual climate modes, the effects of sea-level rise, and changing intensity of human fire usage. This work also pro- vides an improved understanding of the evolution of climate of southeastern Australia and how climate dynamics affect vegetation distributions and landscape fire activity. Late glacial and Holocene vegetation changes and associated drivers on King Island The existing published paleoecological study from King Island (Egg Lagoon) suggests the prevalence of herbaceous vegetation during the last glacial maximum and late glacial, which is in line with the broader Bass Strait and southeast Australian regional pale- oecological data (D’Costa et al., 1993; Adeleye et al., 2021d). This is supported by our results from Lake Martha Lavinia, which show a Poaceae- (grass-) dominated landscape during the late glacial between ca. 15 and 13 ka (Fig. 5). A grassy Eucalyptus–Banksia savanna likely prevailed during this period, with low frequency of fire peaks, which is consistent with small or low-intensity fires. The presence of charcoal throughout the late glacial record, despite low amounts of woody biomass in the landscape compared to the Early Holocene, accords with anthropogenic fire management. This fire regime may have particularly favored serotinous Banksia seed dis- persal and its population expansion in the landscape (Enright and Lamont, 2006). A turnover in vegetation occurred from ca. 13 ka, led by Eucalyptus–Melaleuca forest development as conditions became warmer into the Early Holocene. Lake Martha Lavinia water levels remained generally stable (Adeleye et al., 2025), suggest- ing that moisture increase in the Early Holocene was likely due to reduced evaporation (Wilkins et al., 2013) and not nec- essarily an increase in precipitation. The forest likely became increasingly moist from ca. 11 ka when Acacia expanded (Fig. 5). Acacia presently occupies wet and damp areas on King Island (King Island Natural Resource Management Group, 2002), and in the broader Tasmanian region Acacia co-occurs with Eucalyptus and Melaleuca in wet sclerophyllous forests and swamp forests (Reid et al., 1999; Howells, 2012). High charcoal influx suggests Anthropogenic burning persisted throughout this time, as large Holocene vegetation dynamics, Bass Strait, Tasmania 63 Table 1. AMS radiocarbon ages of bulk sediment, with Calibrated median age ranges at 95% confidence interval for Lake Martha Lavinia, King Island. Sites LabID 14C age (yr BP) Error (1σ) Depth (cm) Median Calibrated age (cal yr BP) Calibrated age range (cal yr BP) Lake Martha Lavinia D-AMS 043097 963 28 34 844 771−920 Lake Marthia Lavinia D-AMS 043098 3321 22 72 3490 3454−3570 Lake Martha Lavinia D-AMS 043099 4405 25 125 4931 4867−5041 Lake Martha Lavinia D-AMS 043100 5326 25 175 6080 5997−6189 Lake Martha Lavinia D-AMS 043101 6414 32 225 7303 7259−7421 Lake Martha Lavinia D-AMS 043102 11,654 48 275 13,480 13,435−13,591 0 50 100 150 200 250 300 0 50 00 10 00 0 15 00 0 Depth (cm) ca l y r B P 0 1500 3000 −5 0 −4 0 −3 0 Iteration 0 300 600 Acc. rate (yr/cm) acc.shape: 1.5 acc.mean: 50 0.0 0.4 0.6 Memory mem.strength: 10 mem.mean: 0.5 62 5 cm sections −5 0 −4 0 −3 0 Lo g of O bj ec tiv e −5 0 −4 0 −3 0 0 0.02 0.04 cm/yr Figure 2. Left: Lake Martha Lavinia Bacon age–depth model and sediment accumulation rates. Right: Markov chain Monte Carlo iterations (top panel), prior (red curves), and posterior distributions (gray histograms) of accumulation rate (middle panel) and its memory (bottom panel). charcoal fragments have been found in Cataraqui Monument Quarry on King Island dating to ca. 12 ka (Bowdler, 2015). The state of the oceanic surface water temperatures around King Island can be traced using records of the tropical planktic foraminifera Globigerinoides ruber (a proxy for Leeuwin Current changes) and alkenone sea surface temperature (SST) records from offshore western Victoria∼200 km from King Island (Moros et al., 2021). The Leeuwin Current (LC) presently transports shallow tropical warm waters especially from Indonesia, through Western Australia’s coast to southern Australia, and as far as western Tasmania, bringing rainfall to these regions. The decline in trop- ical planktic foraminifera G. ruber from ca. 7.5 ka suggests weak LC strength, less-warm waters, and reduced moisture near King Island (Moros et al., 2021). This is also reflected in the sharp decline in Lake Martha Lavinnia water levels at this time (Fig. 5). The LC may have been stronger in South Australia’s waters (Moros et al., 2009) but unlikely as far as the Bass Strait area at this time (Perner et al., 2018; Moros et al., 2021). The onset of an extended dry period on King Island at ca. 7.5 ka was fol- lowed soon thereafter by high fire activity from ca. 7.2 cal ka, peaking at ca. 6.9–5.5 cal ka, with high vegetation turnover occur- ring from ca. 6 cal ka and peaking at ca. 5.5 cal ka The timing of events and the lack of archaeological evidence of human occupa- tion during this period (Bowdler, 2015) suggest that the drier phase on King Island between 7.5 and 5.5 ka drove large fires that sig- nificantly (SCD > 0.2) transformed the pre-existing moist forest (Fig. 5). The IOD also strongly correlates with multi-decadal rainfall variability on King Island today and we suspect its major contri- bution to the Middle Holocene dry period on the island. There is currently no record of Holocene IOD dynamics in Australia; however, a coral record from east Africa has documented the formation of warm waters in the region from 8–5 ka (Leupold et al., 2023). Based on the current pattern of the IOD influence over Australia, warm waters and more rainfall in east Africa in the Middle Holocene possibly meant cool water concentration near Australia with reduced rainfall (positive IOD phases), espe- cially in landmasses in or bordering the eastern Indian Ocean Basin. Reduced rainfall on King Island in the Middle Holocene would have increased biomass dryness, making the moist forest 64 M.A. Adeleye et al. 0 1000 2000 3000 4000 5000 6000 7000 8000 9000 10000 11000 12000 13000 14000 15000 M od el le d ag e (c al yr BP ) 20 40 60 80 100 120 140 160 180 200 220 240 260 280 300 D ep th (c m ) 20 B an ks ia 20 40 E uc al yp tu s 20 40 60 M el al eu ca 20 A ca ci a 20 40 60 Le pt os pe rm um 20 A ll o ca su ar in a cf .m on il i f e ra P om ad er ri s D od on ae a vi sc os a 20 M on ot oc a 20 G oo de ni a 20 40 60 Po ac ea e 20 As te ro id ea e G le ic he ni a m ic ro ph yl la 20 D ic k s on ia an ta rc tic a B le ch nu m P t e rid iu m es cu l e nt um Trees and shrubs FernsHerbs 0 50 100 150 0 1 2 3 %Pollen SCD/500 yrs Taxa/minimum count Macrocharcoal influx (Pieces/cm²/yr) Fire frequency (fires/500 yrs) (a) Lake Martha Lavinia terrestrial pollen record 0 20 400.0 0.2 0.4 Vegetation turnover Floristic richnessM aj or v eg et at io n ph as es LML-1 LML-2 LML-3 LML-4 0 1000 2000 3000 4000 5000 6000 7000 8000 9000 10000 11000 12000 13000 14000 15000 Ag e (c al yr BP ) 40 R es tio na ce ae C yp er ac ea e Tr ig lo ch in P ed ia st ru m Zy gn em a D eb ar ya S pi ro gy ra H dV -1 2 H el ic oo n Zy gn em at ac ea e M ic ro t h y r ia c e ae G lo m us C op ep od a A ss u l in a 300 1500 2000 20 202040 102020 High 0.02.0 Wetland DCA1 Low W at er le ve l ch an ge s NPPs/cm³x100 Wetland plants Freshwater algae Fungal spores Zoological (b) Lake Martha Lavinia wetland pollen and non-pollen palynomorph record LML-1 LML-2 LML-3 LML-4 M aj or v eg et at io n ph as es Figure 3. Lake Martha Lavinia terrestrial pollen and charcoal records (a) from this study and (b) wetland taxa record from Adeleye et al. (2025). Red-dashed line in panel (a) is the empirical threshold (SCD = 0.2) for significant turnover previously derived for southeast Australia (Adeleye et al., 2021d). SCD > 0.2 means complete shift in pollen assemblages, which is used to identify major phases of vegetation communities in the pollen record. Black curve in panel (b) is the loess-smoothed first axis of Detrended Correspondence Analysis (DCA) of wetland taxa, with bootstrapping at 95% confidence interval (gray-shaded area). susceptible to burn in events of lightning, leading to forest opening. Lightning strikes are common on the island today (e.g., 6000- ha area burned in lightning-caused fires during the summer of 2001; Corbett, 2010). During the 8–5 ka interval, the forest under- story became more open, particularly with the exclusion of Acacia trees and understory dominance by disturbance-tolerant shrubs (Monotoca) and tree ferns (D. antarctica). Monotoca is a common understory shrub in Melaleuca, Eucalyptus, and Acacia forest com- munities with an absence of a dense understory, and D. antarctica usually occupy forest canopy gaps on King Island (King Island Natural Resource Management Group, 2002). While direct climate effects (drought and lightning) may explain the Middle Holocene fires and forest transformation at Lake Martha Lavinia, it is important to also note the possible role Holocene vegetation dynamics, Bass Strait, Tasmania 65 Big Reedy Lagoon, Cape Barren Island Lake Martha Lavinia, King Island Divergence index 0.1 0.2 0.3 0.4 m12 squared = 0.58 Correlation = 0.65 p-value = 0.001 0 ka 2 ka 4 ka 6 ka 8 ka 10 ka 12 ka 14 ka −0.2 0.0 0.2 0.4 −0.25 0.00 0.25 0.50 DCA Axis 1 D C A Ax is 2 Figure 4. Detrended Correspondence Analysis (DCA) and Procrustes analyses results, comparing Lake Marth Lavinia, King Island and Big Reedy Lagoon, Cape Barren Island (Truwana). Age labels are sample ages at 2000-yr bins. Dark/black lines indicate high divergence in samples and light/gray lines indicate sample convergence or low divergence. 0 1000 2000 3000 4000 5000 6000 7000 8000 9000 10000 11000 12000 13000 14000 15000 Ag e (c al yr BP ) 20 40 60 80 100 Euc aly ptu s Mela leu ca Ban ks ia Aca cia Mon oto ca Le pto sp erm um Poa ce ae Dick so nia an tar cti ca 20 40 60 80 100 Euc aly ptu s Mela leu ca Cas ua rin a Pom ad err is Epa cri s Spre ng eli a Le pto sp erm um Mon oto ca Poa ce ae 0 50 100 150 Macrocharcoal influx (Pieces/cm²/yr) 0 10 20 Fire High 0.02.0 Low Wetland taxa DCA1 scores G. ruber% 105 LC near Bass Strait 15 12 14 16 Alkenone SST (⁰C) SST near Bass Strait 0.01.02.0 Wetland water level Dry Wet %Pollen 2010 30 No. of El Niño events Dry Wet ENSO activity Max. sea level Vegetation Lake Martha Lavinia, King Island Big Reedy Lagoon, Cape Barren Island La ke M ar th a La vi ni a Bi g R ee dy L ag oo n La ke M ar th a La vi ni a Bi g R ee dy L ag oo n Onset of KI drought Onset of fire on KI Vegetation change on KI Figure 5. Synthesis of vegetation, fire, and wetland changes on King Island and Cape Barren Island (Truwana) alongside nearby regional sea-surface temperature (SST) and Leeuwin Current (LC) record from marine core SS02-06-GC15 offshore western Victoria (De Deckker et al., 2020; Moros et al., 2021; De Deckker, 2022) and the record of El Niño events (Moy et al., 2002). Wetland water-level changes (black curve) is inferred from loess-smoothed first axis of Detrended Correspondence Analysis (DCA) of wetland taxa, with bootstrapping at 95% confidence interval (gray-shaded area). Blue horizontal line indicates the onset of dry episode on King Island at ca. 7.5 cal ka, followed by increased fires from ca. 7.2 cal ka (solid black line) and major vegetation turnover from ca. 6 cal ka (green line). Blue arrows also point to the cooling and drying trends reflected by SST, LC, and wetland water-level changes at ca. 7.5 cal ka, while black arrows indicate the increase trend of fire from ca. 7.2 cal ka. Black-dashed line is the timing of sea-level transgression in southeast Australia at ca. 6.9 cal ka (Dougherty et al., 2019). of sea-level change. Sea level reached a maximum in southeast Australia at ca. 6.9 ka (Dougherty et al., 2019), which corresponds with the timing of the expansion of Monotoca, which is also found in coastal scrub on King Island. Sea-level rise effects, especially salt spray and wind disturbance, as the island coastline contracted also may have contributed to the shrub expansion, particularly in locations proximal to the coast in the Middle Holocene (Fig. 5). The Middle Holocene vegetation at Lake Martha Lavinia possi- bly would have been a mix of open forest and scrub communities, which generally continued into the Late Holocene, with some internal community turnover. The abundance of Melaleuca and Leptospermum at ca. 5.5–4 ka likely reflects middle successional stage forest development, with tall shrubby layers (King Island Natural Resource Management Group, 2002) as forest started closing again after the Middle Holocene dry period and fire disturbance. Eucalyptus likely dom- inated the forest canopy after 4 ka with more diverse understory shrubs (e.g., Pomaderris, Monotoca) as the climate became wet- ter (Fig. 5). However, the driver of Late Holocene wet conditions remains unclear. The LC and SST dynamics points to a general drying and cooling trend during this period, while increasing wet- land water levels suggest a wetting trend on the island. Given the 66 M.A. Adeleye et al. dominance of IODandSAMon the island today, the LateHolocene climate may have been dominated by negative IOD phases and strong westerlies influence, resulting in increased precipitation. Long-term vegetation dynamics on western vs eastern Bass Strait islands The paleoecological record from Cape Barren Island shows herba- ceous vegetation also characterized the eastern Bass Strait Island during the late glacial before 13 ka (Adeleye et al., 2021b; Fig. 5). Woody communities dominated by Eucalyptus and Melaleuca expanded from 13 ka as well, with charcoal evidence of fires start- ing at ca. 12 ka. Wet-habitat and tall forest Eucalyptus species (e.g., E. globulus) are thought to have originally occurred onCapeBarren Island from 13 ka as climate became warmer and wetter. Frequent fires derived from Aboriginal burning practices from ca. 12 ka favored fire-adapted and dwarf woodland Eucalyptus species (e.g., E. nitida) at the expense of the pre-existing forest (Adeleye et al., 2023b). The earlier decline of Eucalyptus on Cape Barren Island by ca. 9 ka in favor of ericaceous coastal heathland communities has been linked to the effect of sea-level rise on the relatively small island, with forest cover limited by recurrent wind breakage and salt stress (Adeleye et al., 2021b). While Eucalyptus also started to decline from ca. 9 ka on the larger Flinders Island, ∼8 km north of Cape Barren Island, the decline was slower, with Eucalyptus still dom- inating vegetation on the island until ca. 6.5 ka (Fig. 6), which is shortly after sea level transgression ceased in southeast Australia (Dougherty et al., 2019). Eucalyptus tree cover declined in favor of Casuarinaceae heathland expansion on Flinders Island (McWethy et al., 2017) and the coast of northeast Tasmania (∼50 km south of Cape Barren Island) at this time (Thomas and Kirkpatrick, 1996). Wetland water levels on Cape Barren Island also declined from ca. 7 ka, with low levels sustained in the Late Holocene as ENSO intensified (Adeleye et al., 2021d). Diverse coastal heathland and scrub communities were maintained on the island through the Late Holocene, likely due to generally high sea levels and low fire activity (Adeleye et al., 2021b). Drought-tolerant Callitris trees also increased, especially on the larger Flinders Island (McWethy et al., 2017), which reflects widespread ENSO-driven aridity across eastern Australia during this period (Barr et al., 2019). The DCA and Procrustes analyses suggest the greatest similar- ity (low divergence) in vegetation communities was during the late glacial to Middle Holocene (14–6 ka; correlation = 0.65, p-value = 0.001; Figs. 5 and 6). Vegetation development on both western and eastern Bass Strait islands during this period tracks the post- glacial gradual warming and wetting trend, with a transition from herbaceous to wet woodland or forest. Marked divergence in vegetation trajectories occurred from the Middle Holocene. While relative dryness and fire were major drivers of forest transformation on King Island at this time, sea- level rise was the dominant factor of terrestrial vegetation changes on Cape Barren Island, with a major expansion of treeless coastal heathland communities. Meanwhile sea-level rise effect was likely minimal on the larger King Island as evinced by the stability of Eucalyptus tree abundance (Fig. 5). The increase in Monotoca and Pomaderris on both islands from the Middle Holocene can be linked to the effect of sea-level rise, given the tolerance of both species to salt spray and adaptation to soil disturbance (King Island Natural Resource Management Group, 2002). Nonetheless, diver- gence in vegetation is reflected in the typical coastal heathland and scrub communities that persisted on Cape Barren Island through the Middle to Late Holocene, while forest continued to be a dom- inant vegetation component on King Island (Fig. 6). This suggests direct oceanic influence continued to be important in shaping ter- restrial vegetation assemblages on the smaller Cape Barren Island in the Late Holocene while climate control remained dominant in shaping vegetation on the larger King Island. Climate dominance likely resumed on the large Flinders Islands near Cape Barren Island as well in the Late Holocene (McWethy et al., 2017). Insights into southeast Australia’s paleoclimate Existing studies of southeast Australian paleoclimate and pale- oclimatic effect on vegetation and wetland ecosystems generally agree on the increasing influence of ENSO in many parts of the region during the Middle to Late Holocene (Fletcher and Moreno, 2012; Beck et al., 2017; McWethy et al., 2017; Barr et al., 2019; De Deckker, 2022). By contrast, Early to Middle Holocene climate across the region has been characterized as gradually warming and wetting, as Southern Westerlies (SW) strengthened, reaching a zenith by the Middle Holocene (Wilkins et al., 2013; De Deckker, 2022). A synthesis study using various data streams suggests a geographically heterogeneous hydroclimatic pattern across south- east Australia during the Early to Middle Holocene (Fletcher and Moreno, 2012) driven by fluctuations in the strength and direction of SW-modulated precipitation gradient.This climate anti-phasing is clearly reflected in plant functional trait responses (e.g., leaf and seed traits) in the region during the Holocene (Adeleye et al., 2023a), and this climatic pattern is thought to have also prevailed, especially in Tasmania, during the early, middle, and late glacial periods (Fletcher and Thomas, 2010; Fletcher and Moreno, 2012; Adeleye et al., 2024). Our results contrast with previous findings of east–west cli- mate anti-phasing controlled by SW dominance during the Early to Middle Holocene in southeast Australia (Fletcher and Moreno, 2012) and peak wetness in the Middle Holocene across the region (De Deckker, 2022). Given the well-situated locations of King Island and Cape Barren Island in western and eastern sectors of southeast Australia, respectively, we expect the paleoecology of these islands to reflect the late glacial–MiddleHolocene SW-driven climate anti-phasing. However, late glacial–Middle Holocene veg- etation onKing Island andCape Barren Island shows a clear transi- tion fromdry, open herbaceous towoodland or forest communities (Figs. 5 and 6) that tracks the general post-glacial warming and wetting trend (Wilkins et al., 2013; De Deckker, 2022), with sea- level rise and human land use driving localized dynamics. It is possible that the SW-driven climatic anti-phasing followed a lat- itudinal gradient in southeast Australia after termination of the last glaciation, with a stronger signal in the higher latitude area of Tasmania than in the Bass Strait and southeast Australian main- land. This may explain why the climate anti-phasing signal was more evident in the vegetation composition and fire regimes in Tasmania (Fletcher andMoreno, 2012;Mariani and Fletcher, 2017) and not in the more northerly Bass Strait Islands. The Middle Holocene (8–5 cal ka) is thought to be a period of maximum warmth and wetness across southeast Australia (Wilkins et al., 2013; De Deckker, 2022), with the SW bringing more rainfall to western southeast Australia and a reduction in fire activity in western Tasmania at this time (Fletcher and Moreno, 2012). Therefore, we also expected fire activity to be low on King Island in the Middle Holocene, especially given the current lack of archaeological evidence of human occupation at this time, but this was not the case. Fire activity was in fact greatest on the island Holocene vegetation dynamics, Bass Strait, Tasmania 67 Ag e (c al y r B P) Lake Martha Lavinia, King Island Middle Patriarch, Flinders Island Big Reedy Lagoon, Cape Barren Island Early Holocene Eucalyptus- Melaleuca forest/woodland Late glacial grassy vegetation Mix heathland and scrub Ericaceae heathlandEarly-Mid Holocene Eucalyptus forest/woodland Callitris-Eucalyptus- Casuarince woodland and scrub Casuarinace heathland Early Holocene Eucalyptus-Melaleuca initial forest Late seral Acacia- Eucalypt-Melaleuca wet tall forest Mix forest and scrub 1000 2000 3000 4000 5000 6000 7000 8000 9000 10000 11000 12000 13000 14000 0 15000 Eastern Bass StraitWestern Bass Strait Early seral shrubby forest Late glacial grassy vegetation Mix forest and scrub W et D ry W et D ry D ry W et D ry Dryness across Bass Strait Figure 6. Summary of vegetation and climate trajectories of the Bass Strait region based on pollen records from the three largest islands: King Island, Cape Barren Island (Truwana), and Flinders Island. during this period, accompanied by major forest transformation (Fig. 3a). Relatively dry wetlands (Adeleye et al., 2025) on King Island combinedwith evidence of weak LC (low rainfall), SST cool- ing∼200 km fromKing Island, and potentially positive IODphases at this time indicate a Middle Holocene dry episode, at least in western Bass Strait and potentially across the Bass Strait (Fig. 6). ENSO influence was likely more important in eastern Bass Strait at this time, persisting into the LateHolocene. A complex climate sys- tem that favored increased rainfall, potentially including negative IOD phases and strong SW influence, is likely to have prevailed on King Island in the Late Holocene. Conclusion The new paleoecological record from King Island (western Bass Strait) combined with the existing one from Cape Barren Island (eastern Bass Strait) presented in this study provide an improved understanding of the complexity of the biogeographical and pale- oclimatic history of the current Bass Strait islands and submerged Bassian Plains landscape. The spatio-temporal dynamics of floris- tic composition and drivers of change offer new insights into the paleoclimate of the region and the broader southeast Australia. Late glacial–Holocene terrestrial vegetation in the Bass Strait area has generally followed a similar trajectory, especially during the late glacial–Middle Holocene, although the drivers of change diverged in the Middle Holocene, triggering a corresponding tra- jectory divergence between west and east Bass Strait islands. The effect of Middle Holocene sea-level rise was regional, but was greatest on the eastern islands, with the replacement of wood- land by coastal heathland communities. A relatively drier period accompanied by major fires transformed forest to forest–scrub communities in western Bass Strait at the same time. Climate and high sea levelsmaintainedLateHolocene vegetation in easternBass Strait while vegetation in western Bass Strait was largely shaped by climatic influence. Vegetation trajectories on both King Island and Cape Barren Island suggest a homogeneous warming andwetting climatic trend prevailed over the Bass Strait area and potentially mainland south- east Australia during the late glacial–EarlyHolocene.The influence of SW-driven climatic anti-phasing was possibly greatest in the higher latitude of Tasmania. Although conditions may have been wet in some areas of southeast Australia and Tasmania, a combi- nation of climatic events, including weak LC, positive IOD phases, and the onset of ENSO, likely resulted in a dry episode in Bass Strait during the Middle Holocene. The influence of strong SW and negative IOD phases possibly modulated western Bass Strait climate in the Late Holocene, bringing more rainfall to the region, while ENSO intensification drove rainfall declines in eastern Bass Strait. Acknowledgments. Research on King Island is supported by the Aboriginal Land Council of Tasmania and the Tasmanian Aboriginal Centre. We thank the Working on Country, pakana rangers, and Truwana rangers for guidance and advice while on their country. Permission to core wetlands in the Lavinia State Reserve was granted through the Department of Natural Resources and Environment, Tasmania (No. E22264). Many thanks to the editors and reviewers, especially John Tibby, for his constructive feedback that signifi- cantly improved the manuscript. This research was made possible through an Australian Research Council Centre of Excellence for Australian Biodiversity and Heritage support grant CE170100015 (S. Haberle) and Australian Nuclear Science and Technology Organization grant AP13238 (D. Bowman). D. Bowman also acknowledges Laureate grant from the Australian Research Council (FL220100099) and M. Adeleye acknowledges grant from the UK Research and Innovations Future Leaders Fellowship (MR/Y018176/1). Conflict of interest declaration. Authors declare no conflict of interest. 68 M.A. Adeleye et al. References Australian Bureau of Meteorology, undated. Australian Climate Influences [WWW Document]. http://www.bom.gov.au/watl/about-weather-and- climate/australian-climate-influences.shtml?bookmark=introduction [accessed 13 May 2022] Adeleye, M., Haberle, S., Harris, S., Connor, S., Stevenson, J., 2021a. Assessing long-term ecological changes in wetlands of the Bass Strait Islands, Southeast Australia: palaeoecological insights and management implica- tions. Wetlands 41, 88. https://doi.org/10.1007/s13157-021-01480-z Adeleye,M.A.,Haberle, S.G.,Harris, S.,Hopf, F.V.-L.,Connor, S., Stevenson, J., 2021b. Holocene heathland development in temperate oceanic Southern Hemisphere: key drivers in a global context. Journal of Biogeography 48, 1048–1062. https://doi.org/10.1111/jbi.14057 Adeleye, M.A., Mariani, M., Connor, S., Haberle, S.G., Herbert, A., Hopf, F., Stevenson, J., 2021c. Long-term drivers of vegetation turnover in Southern Hemisphere temperate ecosystems. Global Ecology and Biogeography 30, 557–571. Adeleye, M.A., Haberle, S.G., McWethy, D., Connor, S.E., Stevenson, J., 2021d. Environmental change during the last glacial on an ancient land bridge of southeast Australia. Journal of Biogeography 48, 2946–2960. https: //doi.org/10.1111/jbi.14255 Adeleye, M.A., Haberle, S.G., Gallagher, R., Andrew, S.C., Herbert, A., 2023a. Changing plant functional diversity over the last 12,000 years pro- vides perspectives for tracking future changes in vegetation communities. Nature Ecology & Evolution 7, 224–235. https://doi.org/10.1038/s41559-022- 01943-4 Adeleye, M.A., Haberle, S.G., Hopf, F., Harris, S., McWethy, D.B., 2023b. Insights into the indigenous-managed landscape in southeast Australia dur- ing the Holocene. Vegetation History and Archaeobotany 32, 419–427. https: //doi.org/10.1007/s00334-023-00918-0 Adeleye, M.A., Hopf, F., Haberle, S.G., Stannard, G.L., Mcwethy, D.B., Harris, S., Bowman, D.M.J.S., 2024. Landscape burning facilitated Aboriginal migration into Lutruwita/Tasmania 41,600 years ago. Science Advances 10, eadp6579. https://doi.org/10.1126/sciadv.adp6579 Adeleye, M.A., Haberle, S.G., Hua, Q., Bowman, D.M.J.S., 2025. Palaeoecological analyses reveal recent fires have destroyed Late-Holocene peat deposits in Tasmania’s largest ramsar reserve. Austral Ecology 50, e70087. https://doi.org/10.1111/aec.70087 Barr, C., Tibby, J., Leng, M.J., Tyler, J.J., Henderson, A.C.G., Overpeck, J.T., Simpson, G.L., et al., 2019. Holocene El Niño–Southern Oscillation vari- ability reflected in subtropical Australian precipitation. Scientific Reports 9, 1627. https://doi.org/10.1038/s41598-019-38626-3 Beck, K.K., Fletcher, M.-S., Gadd, P.S., Heijnis, H., Jacobsen, G.E., 2017. An early onset of ENSO influence in the extra-tropics of the southwest Pacific inferred from a 14, 600 year high resolutionmulti-proxy record fromPaddy’s Lake, northwest Tasmania. Quaternary Science Reviews 157, 164–175. https: //doi.org/10.1016/j.quascirev.2016.12.001 Birks, H.J.B., Felde, V.A., Bjune, A.E., Grytnes, J.-A., Seppä,H., Giesecke, T., 2016. Does pollen-assemblage richness reflect floristic richness? A review of recent developments and future challenges. Review of Palaeobotany and Palynology 228, 1–25. https://doi.org/10.1016/j.revpalbo.2015.12.011 Blaauw, M., Christen, J.A., Lopez, M.A.A., Vazquez, J.E., Gonzalez, O.M., V., Belding, T., Theiler, J., Gough, B., Karney, C., 2022. Package ‘rba- con’ Age-DepthModelling using Bayesian Statistics. https://cran.r-project.org/ web/packages/rbacon/rbacon.pdf Bowdler, S., 2015. The Bass Strait Islands revisited. Quaternary International 385, 206–218. https://doi.org/10.1016/j.quaint.2014.07.047 Connor, S.E., van Leeuwen, J.F.N., van der Knaap, W.O. (Pim), Akindola, R.B., Adeleye, M.A., Mariani, M., 2021. Pollen and plant diversity relationships in a Mediterranean montane area. Vegetation History and Archaeobotany 30, 583–594. https://doi.org/10.1007/s00334-020-00811-0 Corbett, K., 2010. Post-Fire Geomorphology and Vegetation Assessment Lavinia State Reserve, King Island, Report 1 – Assessment of Peat Deposits, Fire Damage and Drainage Features. Corbett’s Enterprises Report for Cradle Coast NRM, May 2010. Davis, B.A.S., Collins, P.M.,Kaplan, J.O., 2015.The age and post-glacial devel- opment of the modern European vegetation: a plant functional approach based on pollen data. Vegetation History and Archaeobotany 24, 303–317. https://doi.org/10.1007/s00334-014-0476-9 D’Costa, D.M., Grindrod, J., Ogden, R., 1993. Preliminary environmental reconstructions from lateQuaternary pollen andmollusc assemblages at Egg Lagoon, King Island, Bass Strait. Australian Journal of Ecology 18, 351–366. https://doi.org/10.1111/j.1442-9993.1993.tb00462.x De Deckker, P., 2022. The Holocene hypsithermal in the Australian region. Quaternary Science Advances 7, 100061. https://doi.org/10.1016/j.qsa.2022. 100061 De Deckker, P., Moros, M., Perner, K., Blanz, T., Wacker, L., Schneider, R., Barrows, T.T., O’Loingsigh, T., Jansen, E., 2020. Climatic evolution in the Australian region over the last 94 ka – spanning human occupancy –, and unveiling the Last Glacial Maximum. Quaternary Science Reviews 249, 106593. https://doi.org/10.1016/j.quascirev.2020.106593 Department of Natural Resources and Environment Tasmania, 2021. Land Information System Tasmania (the LIST). Land Tasmania, Hobart. https:// nre.tas.gov.au/land-tasmania/the-list [accessed 28 January 2024] Dougherty, A.J., Thomas, Z.A., Fogwill, C., Hogg, A., Palmer, J., Rainsley, E., Williams, A.N., et al., 2019. Redating the earliest evidence of the mid-Holocene relative sea-level highstand in Australia and implications for global sea-level rise. PLoS ONE 14, e0218430. https://doi.org/10.1371/ journal.pone.0218430 Enright, N., Lamont, B., 2006. Fire temperatures and follicle-opening require- ments in 10 Banksia species.Australian Journal of Ecology 14, 107–113. https: //doi.org/10.1111/j.1442-9993.1989.tb01012.x Faegri, K., Iversen, J., 1975. Textbook of Pollen Analysis, 3rd rev. ed. Hafner Press, New York. Fletcher, M.-S., Moreno, P.I., 2012. Have the Southern Westerlies changed in a zonally symmetric manner over the last 14,000 years? A hemisphere-wide take on a controversial problem. Quaternary International 253, 32–46. https: //doi.org/10.1016/j.quaint.2011.04.042 Fletcher, M.-S., Thomas, I., 2010. A quantitative Late Quaternary tempera- ture reconstruction from western Tasmania, Australia. Quaternary Science Reviews 29, 2351–2361. https://doi.org/10.1016/j.quascirev.2010.06.012 Higuera, P.E., Brubaker, L.B., Anderson, P.M., Hu, F.S., Brown, T.A., 2009. Vegetation mediated the impacts of postglacial climate change on fire regimes in the south-central Brooks Range, Alaska. Ecological Monographs 79, 201–219. https://doi.org/10.1890/07-2019.1 Higuera, P., Gavin, D., Bartlein, P., Hallett, D., 2010. Peak detection in sediment–charcoal records: impacts of alternative data analysis methods on fire-history interpretations. International Journal of Wildland Fire 19, 996–1014. https://doi.org/10.1071/WF09134 Hogg, A.G., Heaton, T.J., Hua, Q., Palmer, J.G., Turney, C.S., Southon, J., Bayliss, A., et al., 2020. SHCal20 Southern Hemisphere Calibration, 0–55,000 Years cal BP. Radiocarbon 62, 759–778. https://doi.org/10.1017/ RDC.2020.59 Howells,C., 2012.Tasmania’sNatural Flora. AustralianPlants SocietyTasmania Inc., Hobart Group, 431 pp. King IslandNatural ResourceManagementGroup, 2002.King Island Flora: A Field Guide. King IslandNatural ResourceManagement Group, Currie, King Island, Tasmania. Lambeck, K., Chappell, J., 2001. Sea level change through the last glacial cycle. Science 292, 679–686. https://doi.org/10.1126/science.1059549 Leupold, M., Pfeiffer, M., Watanabe, T.K., Nakamura, N., Reuning, L., Blume, A., McClanahan, T., et al., 2023. Mid-Holocene expansion of the Indian Ocean warm pool documented in coral Sr/Ca records from Kenya. Scientific Reports 13, 777. https://doi.org/10.1038/s41598-023-28017-0 Mariani, M., Fletcher, M.-S., 2017. Long-term climate dynamics in the extra- tropics of the South Pacific revealed from sedimentary charcoal analy- sis. Quaternary Science Reviews 173, 181–192. https://doi.org/10.1016/j. quascirev.2017.08.007 Meltsov, V., Poska, A., Odgaard, B.V., Sammul, M., Kull, T., 2011. Palynological richness and pollen sample evenness in relation to local floris- tic diversity in southern Estonia.Review of Palaeobotany and Palynology 166, 344–351. https://doi.org/10.1016/j.revpalbo.2011.06.008 Meltsov, V., Poska, A., Reitalu, T., Sammul, M., Kull, T., 2013. The role of landscape structure in determining palynological and floristic richness. http://www.bom.gov.au/watl/about-weather-and-climate/australian-climate-influences.shtml?bookmark=introduction http://www.bom.gov.au/watl/about-weather-and-climate/australian-climate-influences.shtml?bookmark=introduction https://doi.org/10.1007/s13157-021-01480-z https://doi.org/10.1111/jbi.14057 https://doi.org/10.1111/jbi.14255 https://doi.org/10.1111/jbi.14255 https://doi.org/10.1038/s41559-022-01943-4 https://doi.org/10.1038/s41559-022-01943-4 https://doi.org/10.1007/s00334-023-00918-0 https://doi.org/10.1007/s00334-023-00918-0 https://doi.org/10.1126/sciadv.adp6579 https://doi.org/10.1111/aec.70087 https://doi.org/10.1038/s41598-019-38626-3 https://doi.org/10.1016/j.quascirev.2016.12.001 https://doi.org/10.1016/j.quascirev.2016.12.001 https://doi.org/10.1016/j.revpalbo.2015.12.011 https://cran.r-project.org/web/packages/rbacon/rbacon.pdf https://cran.r-project.org/web/packages/rbacon/rbacon.pdf https://doi.org/10.1016/j.quaint.2014.07.047 https://doi.org/10.1007/s00334-020-00811-0 https://doi.org/10.1007/s00334-014-0476-9 https://doi.org/10.1111/j.1442-9993.1993.tb00462.x https://doi.org/10.1016/j.qsa.2022.100061 https://doi.org/10.1016/j.qsa.2022.100061 https://doi.org/10.1016/j.quascirev.2020.106593 https://nre.tas.gov.au/land-tasmania/the-list https://nre.tas.gov.au/land-tasmania/the-list https://doi.org/10.1371/journal.pone.0218430 https://doi.org/10.1371/journal.pone.0218430 https://doi.org/10.1111/j.1442-9993.1989.tb01012.x https://doi.org/10.1111/j.1442-9993.1989.tb01012.x https://doi.org/10.1016/j.quaint.2011.04.042 https://doi.org/10.1016/j.quaint.2011.04.042 https://doi.org/10.1016/j.quascirev.2010.06.012 https://doi.org/10.1890/07-2019.1 https://doi.org/10.1071/WF09134 https://doi.org/10.1017/RDC.2020.59 https://doi.org/10.1017/RDC.2020.59 https://doi.org/10.1126/science.1059549 https://doi.org/10.1038/s41598-023-28017-0 https://doi.org/10.1016/j.quascirev.2017.08.007 https://doi.org/10.1016/j.quascirev.2017.08.007 https://doi.org/10.1016/j.revpalbo.2011.06.008 Holocene vegetation dynamics, Bass Strait, Tasmania 69 Vegetation History and Archaeobotany 22, 39–49. https://doi.org/10.1007/ s00334-012-0358-y McWethy, D.B., Haberle, S.G., Hopf, F., Bowman, D.M.J.S., 2017. Aboriginal impacts on fire and vegetation on a Tasmanian island. Journal of Biogeography 44, 1319–1330. https://doi.org/10.1111/jbi.12935 Moros, M., De Deckker, P., Jansen, E., Perner, K., Telford, R.J., 2009. Holocene climate variability in the Southern Ocean recorded in a deep- sea sediment core off South Australia. Quaternary Science Reviews 28, 1932–1940. https://doi.org/10.1016/j.quascirev.2009.04.007 Moros, M., Deckker, P.D., Perner, K., Ninnemann, U.S., Wacker, L., Telford, R., Jansen, E., Blanz, T., Schneider, R., 2021. Hydrographic shifts south of Australia over the last deglaciation and possible interhemispheric linkages. Quaternary Research 102, 130–141. https://doi.org/10.1017/qua. 2021.12 Moy, C.M., Seltzer, G.O., Rodbell, D.T., Anderson, D.M., 2002. Variability of El Niño/Southern Oscillation activity at millennial timescales during the Holocene epoch. Nature 420, 162–165. https://doi.org/10.1038/nature01194 Perner, K., Moros, M., De Deckker, P., Blanz, T., Wacker, L., Telford, R., Siegel, H., Schneider, R., Jansen, E., 2018. Heat export from the tropics drives mid to late Holocene palaeoceanographic changes offshore southern Australia. Quaternary Science Reviews 180, 96–110. https://doi.org/10.1016/ j.quascirev.2017.11.033 Oksanen, J., Simpson, G.L., Blanchet, F.G., Kindt, R., Legendre, P.,Minchin, P.R., O’Hara, R.B., et al., 2024. Package ‘vegan’ Community Ecology Package. https://cran.r-project.org/web/packages/vegan/vegan.pdf Reid, J.B., Hill, R.S., Brown, M.J., Hovenden, M.J. (Eds.), 1999. Vegetation of Tasmania. Flora of Australia Supplementary Series Number 8. Australian Biological Resources Study, Environment Australia, Canberra, 456 pp. Sim, R., 1990. Prehistoric sites on King Island in the Bass Strait: results of an archaeological survey. Australian Archaeology 31, 34–43. Sim, R., 1998. The archaeology of isolation?: prehistoric occupation in the Furneaux Group of Islands, Bass Strait, Tasmania. PhD Thesis, The Australian National University, Canberra. https://doi.org/10.25911/ 5d7783717a054 Sim, R.,Thorne, A., 1990.Pleistocene human remains fromKing Island, south- eastern Australia. Australian Archaeology 31, 44–51. Simpson, G.L., Oksanen, J., Maechler, M., 2024. Package ‘analogue’ Analogue and Weighted Averaging Methods for Palaeoecology. https://cran.r-project. org/web/packages/analogue/analogue.pdf Sloss, C.R., Murray-Wallace, C.V., Jones, B.G., 2007. Holocene sea-level change on the southeast coast of Australia: a review. The Holocene 17, 999–1014. https://doi.org/10.1177/0959683607082415 Thomas, I., Kirkpatrick, J.B., 1996. The roles of coastlines, people and fire in the development of heathlands in northeast Tasmania. Journal of Biogeography 23, 717–728. https://doi.org/10.1111/j.1365-2699.1996. tb00032.x Wilkins, D., Gouramanis, C., De Deckker, P., Fifield, L.K., Olley, J., 2013. Holocene lake-level fluctuations in Lakes Keilambete and Gnotuk, south- western Victoria, Australia. The Holocene 23, 784–795. https://doi.org/10. 1177/0959683612471983 https://doi.org/10.1007/s00334-012-0358-y https://doi.org/10.1007/s00334-012-0358-y https://doi.org/10.1111/jbi.12935 https://doi.org/10.1016/j.quascirev.2009.04.007 https://doi.org/10.1017/qua.2021.12 https://doi.org/10.1017/qua.2021.12 https://doi.org/10.1038/nature01194 https://doi.org/10.1016/j.quascirev.2017.11.033 https://doi.org/10.1016/j.quascirev.2017.11.033 https://cran.r-project.org/web/packages/vegan/vegan.pdf https://doi.org/10.25911/5d7783717a054 https://doi.org/10.25911/5d7783717a054 https://cran.r-project.org/web/packages/analogue/analogue.pdf https://cran.r-project.org/web/packages/analogue/analogue.pdf https://doi.org/10.1177/0959683607082415 https://doi.org/10.1111/j.1365-2699.1996.tb00032.x https://doi.org/10.1111/j.1365-2699.1996.tb00032.x https://doi.org/10.1177/0959683612471983 https://doi.org/10.1177/0959683612471983