sign in sign up
<<
Home , Astelia alpina, Baeckea gunniana, Bauera rubioides, Carpha alpina, Epacris serpyllifolia, Eucalyptus coccifera

Review of Palaeobotany and Palynology 233 (2016) 1–11

Contents lists available at ScienceDirect

Review of Palaeobotany and Palynology

journal homepage: www.elsevier.com/locate/revpalbo

Towards understanding the fossil record better: Insights from recently deposited macrofossils in a -dominated subalpine environment

Giselle A. Astorga ⁎, Gregory J. Jordan, Timothy Brodribb

School of Biological Sciences, University of , Private bag 55, , Tasmania 7001, article info abstract

Article history: Accumulations of plant macrofossils in lake sediments and other sedimentary deposits are increasingly being Received 27 January 2014 used to refine our understanding of past vegetation history, ecological processes and related climate conditions. Received in revised form 20 June 2016 However, past vegetation studies based on the use of disarticulated plant structures need to consider the specific Accepted 23 June 2016 potential for fossilisation of different and different plant organs. Such knowledge is available for many Available online 28 June 2016 systems, but the taphonomy of sclerophyll floras is very poorly known. Keywords: To provide understanding of the taphonomic processes affecting the representation of sclerophyllous plant Plant taphonomy species in fossil assemblages this study investigated the potential source vegetation of plant remains extracted Surface sediments from modern sediments of a subalpine lake in Tasmania, southernmost Australia. It was found that the vast Plant macrofossils majority of the types represented in the sediments belong to broadleaf sclerophyllous species living in Plant megafossils close proximity to the lake, although the representation of species was not related to their values of leaf mass Megaflora per unit area. Leaf assemblages Additionally, a bias between the abundance of species in the standing vegetation and the number of of the Sclerophyll vegetation same species in sediments was observed. Thus, small-leaved species, such as many members of , Representation produce comparatively many more leaves and tend to be over-represented in sediments. In contrast, even though, large-leaved species such as and Nothofagus are dominant in the standing vegetation, they produce substantially fewer foliar organs per ground area of vegetation. Accounting for these discrepancies, we developed an intrinsic representativity index that provides a more accurate picture of the relationship between the leaf assemblages incorporated in the sediments and the abundance of these species in the source vegetation. © 2016 Elsevier B.V. All rights reserved.

1. Introduction implications for the interpretation of plant macrofossil records. However, there have been very few such studies. Evergreen sclerophyll floras are widely distributed around the Accumulations of plant macrofossils (also known as megafossils) in world, especially in Mediterranean-type climates, where they represent different depositional environments, such as lakes or streambeds are the most diverse floras outside the tropics (Cowling et al., 1996). used to refine our understanding of past vegetation history, ecological However, plant macrofossil evidence indicates that diverse sclerophyll processes and related climate conditions (e.g. Allen and Huntley, floras existed under wet non-Mediterranean climates in the Cenozoic, 1999; Birks, 2001; Huntley, 2001; Collinson et al., 2010; Gee, 2005). and even as recently as the early , leading to questions about However, plant macrofossil assemblages can only be validly interpreted the link between sclerophylly and dry climates (e.g. Axelrod, 1975; in the light of the potential biases resulting from the differential preser- Chen et al., 2014; Hill, 2004; Palamarev, 1989; Schnitzler et al., vation of different organs and species. The analysis of plant macrofossils 2011; Sniderman et al., 2013). Plant taphonomic studies investigating from surface sediment samples, the recently deposited sediments in the potential for fossilisation of different plant organs and species in depositional environments such as lakes, can enhance the understand- sclerophyll-dominated environments may, therefore, have important ing of processes that determine the differential potential for fossilisation (Dieffenbacher-Krall and Halteman, 2000; Dieffenbacher-Krall, 2007; Spicer and Wolfe, 1987). ⁎ Corresponding author at: School of Biological Sciences (Life Science Building), The potential for fossilisation in may vary depending on both University of Tasmania, Private Bag 55, Hobart, TAS 7001, Australia. E-mail addresses: [email protected] (G.A. Astorga), intrinsic and extrinsic factors (Martin, 1999; Spicer, 1991). In particular, [email protected] (G.J. Jordan), [email protected] (T. Brodribb). intrinsic factors or individual characteristics of plant organs (e.g. the

http://dx.doi.org/10.1016/j.revpalbo.2016.06.004 0034-6667/© 2016 Elsevier B.V. All rights reserved. 2 G.A. Astorga et al. / Review of Palaeobotany and Palynology 233 (2016) 1–11 degree of sclerophylly, number, weight, size and chemical composition macroremains collected by Hill and Gibson (1986), and information of leaves and reproductive structures) can be major determinants of the of present day vegetation within the catchment of the lake. Although likelihood of a species becoming fossilised. This is mainly because these Hill and Gibson (1986) made extensive macrofossil collections, they factors can affect the capacity of plant organs to be transported and made no analyses of reproductive structures or detailed consideration preserved in sediments (Ferguson, 1985, 2005; Gastaldo and Demko, of the potential source vegetation. Thus, this study aims to elucidate 2011; Spicer, 1989, 1991). Additionally, the potential for fossilisation is the modern representation of plant macrofossils within the sediments also affected by extrinsic factors (i.e. the natural characteristics of an of Lake Dobson, the identification of key taphonomic factors likely area). For instance, topographic features of the depositional environ- affecting the final representation of plant species within sediments, ment, the geographic distance of plant communities to the site of and to resolve whether the observed patterns are similar to those deposition, and the presence and ability of wind and/or flowing water found in other systems and geographic areas. to transport plant material (Ferguson, 2005; Gastaldo and Demko, 2011; Greenwood, 1992; Spicer, 1989; Spicer and Wolfe, 1987). Thus, 2. Materials and methods complex interactions of intrinsic and extrinsic factors may determine the final representation of plant parts within sediments resulting in fossil 2.1. Study site assemblages that rarely reflect their source vegetation in simple one to one relationships of abundance (Allen and Huntley, 1999; Birks, 2001, Lake Dobson (42°41′ S, 146° 35′ E) is located in the sub-alpine area 2013; Birks and Birks, 1980; Spicer, 1991; Spicer and Wolfe, 1987). of Mt. Field National Park, south-central Tasmania, Australia at an Plant taphonomic studies can usually provide insights into the altitude of 1034 m. The lake covers c. 5.7 ha, and occupies a glacial cirque relationship between the abundance of organs from a specific taxon bounded on the eastern side by a lateral moraine and on the west part present in the sediments, and the frequency of the same taxon in the by a headwall 250 m high rising at a slope of ~30° (Fig. 1). The lake is source vegetation (Birks, 2013; Birks and Birks, 1980; Spicer and Wolfe, irrigated mainly by two incoming streams: The Golden Stairs Creek 1987). However, in these studies the representativity (i.e. the proportion flowing from the Mawson Plateau down to the east, and Eagle Tarn of organs of a species in terms of the total number of organs preserved in Creek connecting Lake Dobson with Eagle Tarn to the north (Fig. 1). sediments) of any given species may vary markedly depending on the Prevailing winds in Lake Dobson area blow predominantly from west system in which it is measured. Partly because the representativity of a to east (Hill and Gibson, 1986). species is measured relative to the abundance of other represented Most of Tasmania has a temperate climate with mild summers or species, and is therefore non-independent of these other species. Thus, a cool summer in higher plateau areas (Stern et al., 2000). The park is more powerful approach is to use taphonomic studies to develop an located in a transitional area between a perhumid region of wet vegeta- understanding of the potential for fossilisation of different plant organs tion types, including extensive wet sedgeland/heaths and cool temperate and species in depositional settings. closed forests, extending to the west, and a subhumid region of Eucalyptus Taphonomic studies to date, including terrestrial and aquatic open forests and grasslands extending to the east (Harris and Kitchener, systems, especially in Europe (Allen and Huntley, 1999; Birks and 2005; Macphail, 1979; Read, 1999). Annual mean temperature at Lake Birks, 1980; Collinson, 1983; Greatrex, 1983; Spicer, 1981) and North Dobson is 6.2 °C with mean temperature of the warmest and coldest America (Demko et al., 1998; Dieffenbacher-Krall and Halteman, quarter 10.3 °C and 2.3 °C, and annual mean precipitation of 1454 mm. 2000; Dunwiddie, 1987; Spicer and Wolfe, 1987) have identified some Snowfalls are common between July and September. Climatic parameters general taphonomic patterns, although large differences between were estimated using BIOCLIM module from the software package systems and geographic locations also exist. First, such studies suggest ANUCLIM (Xu and Hutchinson, 2013) that combines data from that plant macrofossils are not usually dispersed long distances from meteorological stations and spatial data. their source vegetation. Consequently, plant macrofossil assemblages tend to reflect the local vegetation, with plants growing in distant 2.2. Surveys of vegetation and composition of the standing vegetation communities or allochthonous components being poorly represented in sediments (Birks, 2001, 20,013; Gastaldo et al., 1996; McQueen, Several vegetation types were observed in the catchment of 1969; Rich, 1989; Rowell et al., 2001; Spicer and Wolfe, 1987). Second, Lake Dobson. The alpine vegetation, above the altitudinal tree line, although these investigations have shown that the composition of occurs ~1220 m of altitude, and is mainly composed of a mosaic of species in the standing vegetation broadly mirror to that present in sedgeland, herbfield, cushion plants, sclerophyll heaths and patches forest litter or sediments, the relative contribution of each dominant of dwarf coniferous species including the endemics Diselma archeri, species can be distorted (Schimanski and Bergstrom, 1998). This is Pherosphaera hookeriana and Microcachrys tetragona, and Podocarpus because leaf counts do not accurately predict the relative abundance lawrencei (Crowden, 1999; Harris and Kitchener, 2005; Kirkpatrick, of species in the standing vegetation, particularly if there are large 1982). differences in leaf size (Burnham, 1993; Collinson, 1983; Drake and The subalpine vegetation presents below the treeline and in areas Burrows, 1980; Hill and Gibson, 1986; Spicer et al., 1987; Steart et al., close to the lake form a mosaic of woodlands, heaths and sedgelands. 2005). Additionally, plants species producing a high number of leaves Woodlands are dominated by Eucalyptus species, although stands of per unit of vegetation or plants producing robust leaves (e.g. evergreen montane rainforest are also present around the lake. Heaths are mainly sclerophyll plants with high leaf mass per unit area: LMA) would be dominated by species of the Ericaceae, , and strongly represented in sediments (Steart et al., 2005). This assumption families, while sedgelands are mainly dominated by members is important because it suggests that taxonomic groups containing of the families , Restionaceae, and Gleicheniaceae these plant traits may be better represented than others in sediments (Crowden, 1999; Harris and Kitchener, 2005). (Cornwell et al., 2008; Wright and Cannon, 2001; Wright et al., 2005). For more detail on the composition of species in the surveyed areas Finally, a vast amount of the above-mentioned taphonomic studies of vegetation see Appendix A, and Fig. 4. describe coniferous or deciduous forest systems, but very few comparable In this study, the hydraulic catchment of Lake Dobson is assumed to studies have been undertaken in sclerophyll-dominated floras, which are represent the potential catchment for the leaves, flowers and found major components of vegetation, especially in the Southern Hemisphere in sediments. Four key areas of vegetation within this catchment (Fig. 1) (Box, 2002; Hill and Orchard, 1999; Jordan, 2011; McGlone et al., 2004). were surveyed using 98 sample quadrats and the composition and the This study investigates the taphonomy of Australian sclerophyll- relative abundance (% ground cover) of species recorded. dominated subalpine vegetation in the area of Lake Dobson, south- Percentages of ground cover where transformed to area of ground cover central Tasmania by combining the information provided by plant (squared metres) when necessary. The areas of vegetation were chosen G.A. Astorga et al. / Review of Palaeobotany and Palynology 233 (2016) 1–11 3

Fig. 1. Location of Lake Dobson in Mount Field National Park, south-central Tasmania, Australia, and vegetation types present in the catchment in colours: (1) white = alpine heath; (3) spots = alpine moorland; (4) grey = Eucalyptus woodland; (5) dark grey = Mountain rain forest (dominated by Nothofagus cunninghamii); (6) waves = subalpine moorland (buttongrass). The numbers indicate the surveyed areas of vegetation: 1) Lakeside vegetation, 2) Eagle Tarn Creek vegetation, 3) Golden Stairs Creek vegetation, and 4) Upper catchment vegetation. Details of the composition of each area are given in the Appendix A. 4 G.A. Astorga et al. / Review of Palaeobotany and Palynology 233 (2016) 1–11 to test whether the distance of plant communities to the site of deposi- species were taken. The ground cover of each sample (harvested area) tion, and the presence of transport mechanisms in association to these was measured. Then, LAI was calculated following Eq. (a). fl vegetation areas, such as owing water, may better explain the origin ÀÁ of plant assemblages incorporated within Lake Dobson sediments. The Total dry mass ðÞg =Leaf mass per unit area gm−2 2 LAI ¼ ðaÞ lakeside vegetation (58 plots of 10 m ) was defined as the standing Harvested areaðÞm2 vegetation immediately surrounding the lake in a 5 m wide band. Vege- tation values for this area were adjusted to allow for foliage overhanging LAI was used to calculate the number of leaves per unit of ground N the lake, so some overhanging species may have covers 100%. Two area of vegetation for each species following Eq. (b). areas of riparian vegetation were located along the two main inflowing 2 ÀÁ streams to the lake; Golden Stairs Creek vegetation (30 plots of 25 m ), LAI m2=m2 2 No:leaves per area of vegetation ¼ and Eagle Tarn Creek vegetation (10 plots of 25 m ). Upper catchment Mean leaf area ðÞm2 2 ÀÁ vegetation (10 plots of 25 m ) represents vegetation growing in areas area of ground cover of vegetation m2 at or above the local treeline (~ 1200 m a.s.l.). Additionally, litter samples ðbÞ from the forest floor of the lakeside vegetation were taken in a transect line at 50 m intervals, and the plant material from twenty-one quadrat plots (0.84 m2) was collected. This material was sorted and the compo- The number of leaves per unit of ground area was then used to calcu- sition and abundance of species recorded. Then, in order to identify late the intrinsic representativity for each of the most common (Fig. 1) differences in leaf size we measured the projected area of 50 leaves per species represented in sediments (c). species using a flatbed scanner and the same procedure as in the deter- No: leaves in sediments mination of leaf area of the fresh material. Additionally, the average leaf Intrinsic representativity ¼ ðcÞ size of the species most commonly represented in the forest litter of the No:leaves per area of vegetation lakeside vegetation and leaf assemblages from the lake sediments was investigated using a one-way ANOVA using the R package Rcmdr (Fox, The apparent representativity index for each of the most common 2005, 2007). species was calculated by dividing the abundance of leaves per each taxon in sediments by the abundance (% of ground cover expressed in 2.3. Sediment samples m2) of that species in the vegetation. Representativity indices were calculated for each of the four areas of vegetation (Fig. 1). One hundred and fifty-three surface sediment samples from Lake Dobson collected by Hill and Gibson (1986) were reassessed. The 2.5. Agreement between plant macro-remains and standing vegetation samples were collected from three transects located across the centre of the lake running parallel to the direction of the prevailing winds. The strength of the association between the relative abundance of Each transect was 210 m in long, and separated from each other by each taxon within the lake sediment, the areas of vegetation, litter sam- 10 m intervals. Samples were collected at 1 m intervals from each ples, and leaf traits was measured using Spearman rank-order correlation shore the first 10 m, 2 m intervals the following 10 m, every 5 m the (Sokal and Rohlf, 1995). Specifically, rank-order correlations were used to last 20 m, and at 10 m in the centre of the lake making a total of 51 assess whether relative abundance of species represented in the plant samples on each transect. The samples were sieved through a macrofossils assemblages and the relative abundance of the same 300 μm mesh and only complete leaves and reproductive structures species in the standing vegetation ranked in equal or opposite numerical were recovered and counted. For more details on the sampling and abundance. extraction methodology of the plant material, see Hill and Gibson The same procedure was used to assess whether total leaf area (1986). representation (number of subfossil leaves multiplied by the leaf The plant material was reassessed under a stereomicroscope at 10– area of the same modern species), and number of leaves per ground 40× magnification. Taxonomic identifications to species level were area of vegetation can better predict the observed patterns of species made by comparison with plant reference collections held at the School abundance when compared with leaf counts. Additionally, the average of Plant Science, University of Tasmania, and based on taxonomic plant leaf size of the species most commonly represented in the forest litter descriptions published by Curtis (1963, 1967); Curtis and Morris (1975, of the lakeside vegetation and leaf assemblages was investigated using 1979, 1994); Stevens et al. (2004) and Weiller (1996, 1999). a one-way ANOVA. All analyses were performed using the R package Rcmdr (Fox, 2005, 2007). 2.4. Leaf traits and representativity index 3. Results For each of the most common species within the lake sediments and areas of vegetation we calculated, leaf area, leaf area index, number of 3.1. Composition of the surface sediment samples leaves per unit of ground area of vegetation, leaf mass per area (LMA) in order to investigate whether these traits can explain the composition Approximately 28,000 identifiable macroscopic plant remains (24,460 and abundance of plant macro-remains found within the sediments of leaves and 3338 reproductive structures) were recovered from the Lake Dobson. 153 surface sediment samples of Lake Dobson (Table 1). The number Leaf dry mass per area (LMA g m−2) of modern species was deter- of plant remains per sample ranged from 2 to 1403, with a mean of mined following the protocol of Cornelissen et al. (2003). The leaf area 182 ± 245 (±SD) and a median of 105. All leaf types were identified, of 20 fully exposed sunlight leaves including petioles from five individ- although 138 (4%) of the reproductive structures remained unidentified. uals per species were measured using a flatbed scanner (resolution The plant macrofossil assemblages represent a relatively diverse 1200 dpi) and the image analysis software IMAGEJ (Schneider et al., flora including , ferns, and angiosperms ( and 2012). After area determinations, the plant material was oven-dried at ). Altogether, leaves and reproductive structures 60 °C for 72 h. Then, LMA was calculated by dividing leaf dry mass by revealed the presence of 45 species from 18 families (Table 1). The com- leaf area of the fresh material. The leaf area index (LAI m2/m2), defined bination of leaves and reproductive structures increased the taxonomic as one half of the total leaf area per unit ground surface area (Jonckheere resolution, especially because leaves failed to record herbs and monocot- et al., 2004), was calculated for each of the common species using the yledons, while reproductive organs lacked records of conifers (Table 1; harvesting method (Breda, 2003). Samples from three individuals per Fig. 2). Thus, reproductive structures report 15 species not recorded as G.A. Astorga et al. / Review of Palaeobotany and Palynology 233 (2016) 1–11 5

Table 1 List of species grouped by clade and family and number of organs founded in the surface sediments samples from Lake Dobson.

Group-family Species Leaves Reproductive structures Presence in catchment

Gymnosperms — conifers D.Don 1955 10 Common throughout in fire refuges Athrotaxis selaginoides D.Don 35 – Absent from catchment Athrotaxis laxifolia Hook. 4 – Very rare, few plants around the lake Diselma archeri Hook.f. 21 – Common in alpine areas Phyllocladus aspleniifolius (Labill.) Hook.f. 3 – Rare, isolated plants near lake Pherosphaera hookeriana W. Archer 836 – Common throughout in fire refuges Microcachrys tetragona (Hook.) Hook.f. 43 – Common, especially at high altitude Podocarpus lawrencei Hook.f. 8 – Common at mid and high altitude

Angiosperms — dicotyledons and basal angiosperms Asteraceae pinifolia (Hook.f.) Benth. 128 9 Common, except in alpine areas Ozothamnus rodwayi Orchard 1 – Common Casuarinaceae Allocasuarina zephyrea L.A.S. Johnson – 1 Absent from catchment rubioides Andrews 1748 – Common throughout Ericaceae Acrothamnus montanus (R.Br.) Quinn – 6 Common at mid altitudes straminea R.Br. 36 54 Common throughout serpyllifolia R.Br. 12,205 381 Very common throughout juniperina (J.R.Forst. & G.Forst.) C.M. Weiller 405 186 Common, except in alpine areas Planocarpa petiolaris (DC.) C.M.Weiller – 19 Common at mid to high altitude gunnii Hook.f. – 32 Restricted to boggy areas Richea pandanifolia Hook.f. – 24 Common throughout Hook.f. 159 117 Very common throughout (R.Br.) F.Muell. – 8 Very common throughout Sm. 79 12 Moderately common near lake cunninghamii (DC.) W.M.Curtis 481 79 Common near lake Trochocarpa thymifolia (R.Br.) Spreng. 11 – Very common throughout Haloragaceae Gonocarpus montanus (Hook.f.) Orchard – 50 Moderately common Myrtaceae Hook.f. 939 1449 Common, except in alpine areas Maiden & Blakely 14 161 Common near lake lanigerum (Aiton) Sm. 2400 72 Common near lake Hook.f. – 31 Common near lake Melaleuca squamea Labill. 2 – Moderately common near lake Nothofagaceae Nothofagus cunninghamii (Hook.) Oerst. 1352 – Common up to the tree line Nothofagus gunnii (Hook.f.) Oerst. 2 – Rare Proteaceae revolutus R.Br. 714 1 Very common throughout (R.Br.) Roem. & Schult. 574 – Very common throughout Orites diversifolius R.Br. 103 – Common low to mid altitude Rubiaceae Coprosma nitida Hook.f. 170 219 Very common throughout Rutaceae Nematolepis squamea (Labill.) Paul G.Wilson – 2 Absent? humifusus R.Br. – 4 Common mid to high altitude Tremandraceae Tetratheca procumbens Gunn ex Hook.f. – 2 Rare, isolated plant near the lake Winteraceae (Poir.) A.C.Sm. 21 411 Common, especially at high altitude

Angiosperms — monocotyledons Asteliaceae alpina R.Br. – 12 Common Milligania spp. – 32 Alpine waterlogged areas Cyperaceae alpina R.Br. – 20 Rare, few plants near lake grandis (Labill.) S.T.Blake – 28 Rare, few plants near lake

Pteridophyta (ferns) Gleicheniaceae R.Br. 11 – Common near lake Undet. taxa Three morphotypes – 104 Unknown Total 24,460 3546

leaves, leaves document 16 species not recorded as reproductive organs, Epacris serpyllifolia, a shrub from Ericaceae, dominated the leaf counts, and only 14 species were recorded as both (Table 1; Fig. 3). contributing with 50% of the total number of leaves. Other well- Common tree and shrub species in closer areas of vegetation to the represented species were the Athrotaxis cupressoides (9%), lake were well represented in sediments. For instance, common in /rupestre (8%), Nothofagus cunninghamii both macrofossil assemblages and standing vegetation were Eucalyptus (7%), and (6%), (Table 1; Fig. 4). Additionally, leaf coccifera, Nothofagus cunninghamii, Epacris serpyllifolia, and the conifers assemblages recorded the presence of seven other conifer species Athrotaxis cupressoides and Pherosphaera hookeriana. In contrast, rare from Podocarpaceae and Cupressaceae, the fern Gleichenia alpina, and species in the standing vegetation and forest litter such as Gahnia grandis, three species from Proteaceae (Table 1). These species were absent , Gonocarpus montanus,andPhyllocladus aspleniifolius were from the assemblages of reproductive structures, with the exceptions equally rare in the lake sediments. of Athrotaxis cupressoides and that were marginally Leaf assemblages provided evidence of the presence of 30 plant represented (Table 1). species within ten families (Table 1). Overall, leaf assemblages were The identification of reproductive structures, on the other hand, dominated by species of producing small and high number of revealed the presence of 28 species among 15 families. Three types of leaves, whereas species of producing bigger leaves were underrep- reproductive structure remained with uncertain taxonomic status after resented in sediments (Table B.1). The twelve most common species the analysis, while one type was only identified to the family level among leaf assemblages accounted for 96% of the total number of leaves. (Table 1). The 15 most common species accounted for 94% of the total 6 G.A. Astorga et al. / Review of Palaeobotany and Palynology 233 (2016) 1–11

Fig. 2. Plant families represented as reproductive structures and leaves (*conifers; **dicotyledons herbs; ***monocotyledons) in the sediment of Lake Dobson.

Fig. 3. Plant species represented in the sediment of Lake Dobson as leaves and reproductive structures. Species present only as reproductive structures (open circles), species present only as leaves (black circles), and species present as both leaves and reproductive number of reproductive organs (Fig. 4). The assemblages of reproductive structures (grey circles). structures were dominated by the endemic subalpine tree Eucalyptus coccifera from Myrtaceae (37%), besides the shrubs Epacris serpyllifolia (15%), Tasmania lanceolata (11%), and Coprosma nitida (6%). Reproductive organs also revealed the presence of monocotyledonous and - 3.3. Leaf mass per area (LMA) ous herb species such as Gahnia grandis, Carpha alpina, ,and Gonocarpus montanus, which were undetectable from the leaf assem- The representation of plant species in sediments was not related to blages. The reproductive structures also increased the taxonomic their values of leaf mass per area (LMA). Furthermore, the correlation resolution for one of the most important families in sub-alpine/ analysis revealed a weak and non-significant (rho = 0.05, p-value = alpine Tasmania, the Ericaceae family, expanding the number of 0.84) association between values of LMA and the representation of identified species from seven to twelve (Table 1). species in sediments of Lake Dobson.

3.4. Leaf counts adjusted by area 3.2. Relative abundance Adjusting the number of subfossil leaves in an area basis (leaf counts There were positive correlations between the number of leaves in of individual species multiplied by the mean leaf area of the same the lake sediments and the relative abundance of the same species in species) improved the ability to predict the source vegetation for the standing vegetation for three of the investigated areas of vegetation three of the investigated areas of vegetation compared to leaf counts (Table 2). The strongest relationship was with the vegetation along (Table 2). However, correlation values and significance levels are different Eagle Tarn Creek, although the correlation with the lakeside vegetation from those founded using total plant macrofossil material. In particular, was almost as strong. In contrast, the remaining two areas representing total leaf area representation of the vegetation along the Eagle Tarn the vegetation in the upper catchment (upland vegetation) and the Creek exhibited a much lower and less significant correlation value with vegetation along the Golden Stairs Creek showed a negative and a the leaf assemblages (Table 2). weak positive correlation, respectively (Table 2). The reproductive assemblages showed considerably weaker correla- 3.5. Differential leaf production tion values than did the leaf assemblages (Table 2). In fact, the lakeside vegetation was the only area of vegetation that showed a positive and The shrub species produced many more leaves per ground area significant correlation with the floristic signature of reproductive than the tree species (Table B.1). Particularly obvious was the organs. In contrast, total macrofossil abundance (i.e. sum of leaves and lower number of leaves produced by the Eucalyptus species compared reproductive structures) strongly improved the ability to predict all to Nothofagus cunninghamii and shrub species from Ericaceae (Epacris the areas of vegetation, although the agreement was, once again, partic- serpyllifolia, , Trochocarpa thymifolia), Myrtaceae ularly strong for the lakeside vegetation and the vegetation along the (Leptospermum lanigerum), and Cunoniaceae (Bauera rubioides). More- Eagle Tarn Creek (Table 2). Despite the positive and significant correla- over, the differential productivity of leaves was strongly correlated tion values, it is clear that there are substantial differences between the with the abundance of leaves of the same taxa represented in sediments rank-order sequences of abundance of plant material in sediments and in (Table 2). the standing vegetation, particularly when numbers of leaves are used to reconstruct the relative abundance of species in the potential source veg- 3.6. Litter samples etation. Tree species such as Eucalyptus coccifera, Eucalyptus subcrenulata, and Athrotaxis cupressoides were strongly under-represented in leaf We recovered 5137 leaves from the litter samples of the lakeside assemblages whereas shrub species such as Epacris serpyllifolia, Bauera vegetation recording the presence of twenty-two tree and shrub species rubioides,andOrites acicularis among others were over-represented that represent ~60% of the total number of species, without including (Fig. 4). herbs and ferns, recorded in the surveys of vegetation. Five out of the G.A. Astorga et al. / Review of Palaeobotany and Palynology 233 (2016) 1–11 7

Table 2 Spearman rank correlation values (rho) between leaves, reproductive organs, and total plant representation in the sediment samples of Lake Dobson, and selected areas of the surrounding vegetation (% abundance). P-values = * b 0.05; ** b 0.01; *** b 0.001.

Percentage of relative abundance

Subfossil Reproductive Total sum of No. leaves leaves structures macrofossils adjusted by area

Relative abundance Lakeside vegetation 0.53* 0.44* 0.74*** 0.76*** Golden Stairs Creek 0.18 0.3 0.38* 0.25 Eagle Tarn Creek 0.67*** 0.31 0.63*** 0.54* Upland vegetation −0.02 0.25 0.28 0.31

Number of leaves Lakeside vegetation 0.69*** Golden Stairs Creek 0.25 Eagle Tarn Creek 0.67*** Upland vegetation −0.03

Results of the one-way ANOVA comparing differences in mean leaf size of the most common species represented in the forest litter of the lakeside vegetation and subfossil leaf assemblages showed no signifi- cant differences in mean leaf size among the majority of species in the lake sediments and forest litter. The exception among the eight species evaluated was Leptospermum lanigerum 20% smaller in average area than the forest litter leaves (Fig. 5).

4. Discussion

4.1. Origin of the macrofossil plant material

Several lines of evidence suggest that the plant remains recorded in the sediments of Lake Dobson are likely to be almost exclusively derived from vegetation close to the lake (i.e. lakeside and Eagle Tarn Creek vegetation). Allochthonous elements from distant communities are rare within the surface sediments samples of the lake, although few robust members of the alpine communities, such as the dwarf conifers Microcachrys tetragona and Diselma archeri are represented in small numbers. This input of alpine elements from upper parts of the catchment may have occurred via creek inflow through the Golden Stairs Creek that links the Mawson Plateau with Lake Dobson (Fig. 1). In contrast, common tree and shrub species in closer areas of vegetation are well represented in the sediments, whereas uncommon species in both, standing vegetation and leaf litter (e.g. herbs) are rare in the sediments of the lake. This result supports similar findings by Burnham (1993); Greenwood (1992),and Steart et al. (2005) indicating that rare species in the standing vegetation Fig. 4. Relative abundances of common species in the sediments of Lake Dobson represented as leaves, carpological structures, and the abundance of the same species in will be rare or absent from the forest litter, and therefore, equally rare in areas of vegetation within the catchment. Adjusted abundance: apparent representativity depositional settings. and intrinsic representativity. The intrinsic representativity shows better agreement The limited transportation of the plant remains recorded in the sedi- between the abundance of species in sediments and the lakeside vegetation (see Eq. bandc). ments of Lake Dobson is also suggested by well preserved, and little fragmented leaf assemblages. The vast majority of the leaf types belong to sclerophyllous (broadleaved evergreen) species, which are potentially twenty-two species recorded in the forest-litter were completely absent robust enough to be transported and preserved compared to more fragile from the macrofossil assemblages: , mon- leaf types such as those shed by deciduous species. However, it is also tana,andLomatia polymorpha from Proteaceae, gunniana from know that leaves have only a limited transport and reworking potential Myrtaceae, and pumila from Ericaceae. compared to other plant structures such as pollen (Spicer, 1991). The cover abundance of species in the standing vegetation, and Furthermore, the transport of leaves from more distant communities the number of leaves of the same taxon in litter samples showed a in the catchment of Lake Dobson is limited by the small size of the moderately low and non-significant spearman correlation (rho =0.43, two incoming streams (Fig. 1). p-value = 0.11). However, correcting the number of leaves in sediment Several studies have recognised the distance of the plant community allowing for leaf size resulted in a higher Spearman correlation (rho = to the site of deposition as being critical in the final representation of 0.62, p-value =0.02). plant species in sediments (Ferguson, 1985, 2005; Gastaldo and Demko, The correlation between leaf assemblages and litter samples was 2011; Greenwood, 1991; Spicer, 1989; Spicer and Wolfe, 1987). Results also low and non-significant (rho = 0.44, p-value = 0.10). However, of the correlation analyses also corroborate the same pattern supporting expressing leaf counts in an area basis for litter and subfossil samples, that the distance of the plant community is an effective filter in determin- strongly improved the correlation (rho = 0.68, p-value = 0.01). ing the final representation of plant parts in the sediments of Lake 8 G.A. Astorga et al. / Review of Palaeobotany and Palynology 233 (2016) 1–11

Fig. 5. Size comparison between leaves from the litter and lake sediments. Significance values from ANOVA: 0 = ‘***’,0.001=‘**’,0.01=‘*’.

Dobson. Despite the differences in significance and order of importance in 4.2. Predictors of fossil deposition which the areas of vegetation are predicted by the subfossil material (Table 2), two of the closer areas of vegetation to the lake – the lakeside The number of subfossil leaves per taxon represented in the sediment vegetation and the area of vegetation along Eagle Tarn Creek – are the of Lake Dobson does not predict accurately species abundances in the primary sources of plant material incorporated in the sediments of the standing vegetation, even in terms of rank order (Fig. 4). However, this lake. Thus, species living in close proximity to the site of deposition discrepancy is partially resolved when numbers of subfossil leaves are have more chances to be represented in the sediments than plants adjusted to allow for differences in leaf size (Table 2). On this per leaf living in distant communities (Birks, 2001, 2013; Birks and Birks, 1980; area basis, the agreement between the representativity of dominant McQueen, 1969; Rich, 1989; Rowell et al., 2001; Spicer and Wolfe, 1987). species is particularly strong for the lakeside vegetation. Additional supporting evidence for the local origin of the plant The ability of this correction to improve the agreement between the material is the presence of different plant organs representing the abundance of species in sediments and the abundance of the same same species. This feature is considered to indicate parautochthony or species in the source vegetation can easily be seen by considering the autochthony, mainly because different plant parts may have differing differences in leaf size between the two dominant species in sediments transport potential (Ferguson, 1985, 2005; Gastaldo and Ferguson, and standing vegetation. Epacris serpyllifolia (meanleafareaof4.2mm2) 1998; Gastaldo et al., 1996; Spicer, 1989). Based upon the above- dominates the subfossil assemblages, whereas Eucalyptus coccifera (mean mentioned characteristics it is likely that the mixture of plant parts leaf area of 900 mm2) dominates the lakeside vegetation. An inter- (e.g. leaves, seeds, and delicate floral structures) recorded in the pretation of the vegetation around Lake Dobson based solely on modern sediment of Lake Dobson has undergone limited transport leaf counts from sediments could lead to erroneously conclude that from their source vegetation, and is probably local in origin. Epacris serpyllifolia is the dominant species around the lake. These Similarities in leaf size among litter and sediment also suggest, findings are consistent with those of early studies, showing that despite the small sample size that leaves in the lake sediments are the number of leaves in litter samples does not predict accurately probably derived from plants living in close proximity to the lake. Over- the ranking of species in the standing vegetation particularly when all, the non-significant differences in mean leaf size (Fig. 5) between the there are big differences in leaf size (Burnham, 1993; Collinson, two groups suggest reduced transport and sorting of plant parts. The 1983; Drake and Burrows, 1980; Hill and Gibson, 1986; Spicer and only exception seems to be the shrub Leptospermum lanigerum (Fig. Wolfe, 1987; Steart et al., 2005). 5). It is possible that the reduction in leaf sizes for this species in the lake sediments are due to a moderate transport from areas above the lake (especially the Golden Stairs where its abundance is much higher 4.3. Differential leaf production than other vegetation areas), although this remains elusive because our comparison did not include litter samples from this area of vegetation. The relationship between subfossil leaves and closer areas of vegeta- However, other studies suggest (e.g. Roth and Dilcher, 1978; Greenwood, tion around Lake Dobson significantly improved when using the num- 1992; Steart et al., 2002) that if transportation of plant material occurred ber of leaves that each taxon produce per ground area of vegetation from distant communities, we could expect a reduction in leaf size among compared with abundance data (Table 2). Although this number is the taxa entombed in sediments. Thus, taphonomic biases based on selec- only an approximation of the likelihood for each one of the most com- tion of leaf size due to transportation for systems similar to Lake Dobson mon species to be represented within the sediments, it shows clearly arelikelytobesmall. the differing number of leaves that each species can potentially shed, G.A. Astorga et al. / Review of Palaeobotany and Palynology 233 (2016) 1–11 9 and their relative importance when scaled to their abundance in the a distorted reconstruction of the species dominance in the standing standing vegetation. vegetation (Fig. 4). Apparent representativity index indicates the bias The best examples to explain these considerations are again, Epacris against the preservation of species due to the cost of transport, decay serpyllifolia and the Eucalyptus coccifera. The former is a minor component and differential production in number of leaves. of the vegetation with the higher production of leaves per unit area, Accounting for the differential production of leaves that each species whereas the latter is the dominant tree species in the lakeside vege- may produce, allows for an estimation of the number of leaves that can tation (Fig. 5, and see Table B.1). It is credible that the strong under- be potentially incorporated in sediments. Particularly, the incorporation representation of leaves of Eucalyptus coccifera in sediments compared of specific intrinsic factors in the calculation of representativity (Eq. a to their dominance in the standing vegetation is largely caused by a and b) provided a much more realistic quantification of the contribution combination of taphonomic factors. In particular, the low number of and representation of different tree and shrub species present in the leaves produced per unit ground area (Table B.1), the poor aerial disper- sediments of Lake Dobson (Fig. 4). sion of the leaves limiting their incorporation in sediments, and the The intrinsic representativity index developed in this study indicates restricted preservation potential of the leaves due to their inability to the bias against preservation of a species due to the cost of transport and float (Carpenter and Horwitz, 1988; Hill and Gibson, 1986; Steart et al., decay, and provides a more accurate picture of the relationship of 2005). This result suggest that relying solely in the number of leaves abundance between leaves assemblages in sediments and their source entombed in sediments, without any reference to variation in capacity vegetation (Fig. 4). for leaf production, could lead to erroneous interpretations of fossil records. 5. Conclusions The case of Nothofagus cunninghamii is slightly different. This species appears to be over-represented in sediments once we allow for total leaf The results here show that processes of fossilisation of plant remains area particularly compared to the relative abundance of this species in from sclerophyll-dominated floras may differ quantitatively from typi- the lakeside vegetation. N. cunninghamii has a much lower abundance cal taphonomic processes previously studied mainly in the tropics or in the lakeside vegetation compared to the Eucalyptus species and temperate Northern Hemisphere. This is particularly important because other areas of vegetation although is the fifth more common species modern austral temperate floras differ in taxonomic composition represented in leaf assemblages (Fig. 4). This species also presents a having a much higher proportions of broad-leaved sclerophyllous moderately low production of leaves per ground area of vegetation evergreen species compared to the major deciduous component of the compared to the majority of shrub species, but higher leaf production vegetation that have dominated in northern latitudes (Axelrod, 1966; compared to the Eucalyptus species (see Table B.1). These results are Box, 2002; Jordan, 2011; McGlone et al., 2004) since the Neogene to in agreement with the differential leaf-litter production relatively to the Pleistocene. These differences may have taphonomic implications standing biomass observed by Steart et al. (2005) for N. cunninghamii because thick coriaceous evergreen leaves have different taphonomic and in two different forest systems in Australia. properties (i.e. potential for transportation, buoyancy) than thin papery The differences in production of leaves between these two species leaves of deciduous species (Ferguson, 1985; Spicer, 1981; Steart et al., could also explain why Carpenter and Horwitz (1988) founded that Euca- 2002; Steart et al., 2009). In particular, temperate floras from Australia lyptus obliqua was rarely recorded among litter samples from two creeks exhibit a high diversity of microphyllous and sclerophyllous evergreen in Tasmania. However, the over-representation of N. cunninghamii species, and therefore, plant species producing these leaf traits can be observed by last authors was not associated, in the case of Lake Dobson, strongly represented in fossil deposits. The data from modern sediments to dominance of the species in areas of vegetation close to the lake. of Lake Dobson strongly support some of the taphonomic patterns Furthermore, the abundance of E. coccifera was almost consistently higher observed in northern latitudes studies, although differences between sys- than any other tree and shrub species in three out of four areas of vegeta- tems and geographic locations also exist. Particularly, leaves numerically tion, although the abundance of N. cunninghamii was higher along Eagle dominate the plant assemblages from Lake Dobson over reproductive Tarn Creek. It appears, that different taphonomic factors than those structures. This pattern is consistent with the dominance of leaves and related to abundance in the source vegetation, proximity to the site of the uncommon presence of reproductive structures founded in Cenozoic deposition and rates of leaf production are playing a major role in the fossil floras from Australia (Christophel, 1984; Christophel and Basinger, final representation of these species in the lake sediments. 1982; Hill, 1983; Jordan, 1997; Jordan et al., 1991; Jordan et al., 1995), The representation of plant species in sediments was not predicted and Eocene-Oligocene fossil floras in western North America and by the values of leaf mass per area (LMA) of the species. This result is Europe (DeVore and Pigg, 2010; Mai, 1989; Wing, 1987). However, it remarkable since it is generally accepted that plants producing robust contrasts markedly with some exceptional carpofloras from this period leaves (e.g. high LMA) will be strongly represented in sediments in Europe (e.g. Collinson et al., 2010; Gee, 2005), and Northern (Steart et al., 2005). However, some studies have suggested that high Hemisphere Quaternary fossil deposits (Birks and Birks, 1980) where values of LMA also correlate with high values of leaf life span of species reproductive structures are generally more common than leaves. Despite (Anten, 2002; Reich et al., 1998; Wright and Cannon, 2001). Thus, the dominance of foliar organs in sediment of Lake Dobson, the analysis of species with high LMA may potentially retain foliar organs longer reproductive structures increases the taxonomic resolution, and provides times, and shed leaves at longer intervals that plant species with insight into the relationship between accumulation of plant parts and the lower leaf investments, although this needs to be corroborated with original composition of species in the source vegetation. Particularly, empirical data. reproductive organs may provide information regarding the presence of herbaceous plant species, from which leaves are generally not abscised 4.4. Agreement and representativity of plant macrofossils from the parent plant, and are therefore absent from leaf assemblages. The data from modern sediments of Lake Dobson also support that The apparent representativity index for each of the most common distance of the plant community to the depositional site as one of the species provide insights into the relationship of abundance of organs critical factor affecting the final representation of species in sediments from a specific taxon present in the sediments, and the frequency of (Ferguson, 1985, 2005; Gastaldo and Demko, 2011; Spicer, 1989; the same taxon in the source vegetation (Birks, 2013; Birks and Birks, Spicer and Wolfe, 1987). The floristic composition of plant assemblages 1980; Spicer et al., 1987). However, resulting estimates of the represen- from Lake Dobson strongly agrees with the composition of species in tation of individual species are only relative to other species in the closer areas of vegetation. However, the number of leaves in sediments assemblage. Thus the representativity of a given species may vary mark- does not predict the same abundance of the species in the standing edly depending on the system in which it is measured, and provide also vegetation, mainly because of the large differences in leaf size and rate 10 G.A. Astorga et al. / Review of Palaeobotany and Palynology 233 (2016) 1–11 of production of leaves that exist among tree and shrub species. This Curtis, W.M., 1963. The Student's Flora of Tasmania. Vol. 2. Tasmanian Government Printer, observation is consistent with studies in Australian Nothofagus and Hobart, Australia. Curtis, W.M., 1967. The Student's Flora of Tasmania. Vol. 3. Tasmanian Government Printer, Eucalyptus dominated forests (Steart et al., 2005). Hobart, Australia. Paleoreconstructions of the original forest richness based solely in Curtis, W., Morris, A., 1975. The Student's Flora of Tasmania. Vol 1. second ed. St David's the number of fossil leaves present in the sediments could lead to a Park Publishing, Hobart, Tasmania. Curtis, W.M., Morris, D.I., 1979. The Student's Flora of Tasmania. Vol. 4a. Tasmanian bias towards the over-representation of species producing small leaves, Government Printer, Hobart, Australia. and high amount of foliar organs. Particularly, because shrub species Curtis, W.M., Morris, D.I., 1994. The Student's Flora of Tasmania. Vol. 4b. Tasmanian with small leaves produce much higher number of leaves per unit Government Printer, Hobart, Australia. area than tree species with bigger leaves. Consequently, the chances Demko, T.M., Dubiel, R.F., Parrish, J.T., 1998. Plant taphonomy in incised valleys: implications for interpreting paleoclimate from fossil plants. 26, 1119–1122. that species have of being represented in sediments are potentially DeVore, M.L., Pigg, K.B., 2010. Floristic composition and comparison of middle Eocene to higher if other taphonomic factors are not affecting their deposition late Eocene and Oligocene floras in north america. Bull. Geosci. 85, 111–134. and preservation in sediments. The benefits of including modern infor- Dieffenbacher-Krall, A.C., 2007. Plant Macrofossils Methods and Studies: Surface Samples, Taphonomy, Representation. In: Elias, S.A. (Ed.), Encyclopedia of Quaternary Science. mation of the deposition of plant organs in past plant reconstructions Elsevier, Oxford, pp. 2367–2374. studies may have important implication for the interpretation of fossil Dieffenbacher-Krall, A., Halteman, W., 2000. The relationship of modern plant remains to records. water depth in alkaline lakes in New England, USA. J. Paleolimnol. 24, 213–229. Drake, H., Burrows, C., 1980. The influx of potential macrofossils into Lady Lake, North Supplementary data to this article can be found online at http://dx. Westland, . N. Z. J. Bot. 18, 257–274. doi.org/10.1016/j.revpalbo.2016.06.004. Dunwiddie, P.W., 1987. Macrofossil and pollen representation of coniferous trees in modern sediments from Washington. Ecology 1–11. Ferguson, D.K., 1985. The origin of leaf-assemblages—new light on an old problem. Rev. Acknowledgments Palaeobot. Palynol. 46, 117–188. Ferguson, D.K., 2005. Plant taphonomy: ruminations on the past, the present, and the future. – This paper is based on work done by the main author while the PALAIOS 20, 418 428. Fox, J., 2005. The r commander: a basic-statistics graphical user interface to r. J. Stat. recipient of a CONICYT Bicentennial Becas- Doctoral Scholarship, Softw. 14. and the extra support of the School of Plant Science, University of Fox, J., 2007. Extending the r commander by “plug in” packages. R. News 7, 46–52. Tasmania. The authors thank the Biodiversity Conservation Branch, Gastaldo, R., Demko, T., 2011. The Relationship between Continental Landscape Evolution and the Plant-Fossil Record: Long Term Hydrologic Controls on Preservation. In: DPIPWE for providing the permit to collect native flora in Mount Field Allison, P., Bottjer, D. (Eds.), Taphonomy. Springer, Netherlands, pp. 249–285. National Park. Gastaldo, R.A., Ferguson, D.K., 1998. Reconstructing tertiary plant communities: introductory remarks. Rev. Palaeobot. Palynol. 101, 3–6. Gastaldo, R.A., Ferguson, D.K., Walther, H., Rabold, J.M., 1996. Criteria to distinguish References parautochthonous leaves in tertiary alluvial channel-fills. Rev. Palaeobot. Palynol. 91, 1–21. Allen, J.R., Huntley, B., 1999. Estimating past floristic diversity in Montane regions from Gee, C.T., 2005. The genesis of mass carpological deposits (bedload carpodeposits) in the macrofossil assemblages. J. Biogeogr. 26, 55–73. tertiary of the lower Rhine Basin, Germany. PALAIOS 20, 463–478. Anten, N.P., 2002. Evolutionarily stable leaf area production in plant populations. J. Theor. Greatrex, P., 1983. Interpretation of macrofossil assemblages from surface sampling of Biol. 217, 15–32. macroscopic plant remains in mire communities. J. Ecol. 773–791. Axelrod, D.I., 1966. Origin of deciduous and evergreen habits in temperate forests. Greenwood, D.R., 1991. The Taphonomy of Plant Macrofossils. In: Donovan, S.k. (Ed.), The Evolution 20, 1–15. Processes of Fossilization. Columbia University Press, pp. 141–169. Axelrod, D.I., 1975. Evolution and biogeography of Madrean–Tethyan sclerophyll vegetation. Greenwood, D.R., 1992. Taphonomic constraints on foliar physiognomic interpretations Ann.Mo.Bot.Gard.280–334. of late cretaceous and tertiary palaeoclimates. Rev. Palaeobot. Palynol. 71, 149–190. Birks, H.H., 2001. Plant Macrofossils. In: Smol, J.P., Birks, H.J.B., Last, W.M. (Eds.), Tracking Harris, S., Kitchener, A.E., 2005. From Forest to Fjaeldmark: Descriptions of Tasmania's Environmental Change Using Lake Sediments. Klumer Academic Publisher, Dordrecht, Vegetation. Dept. of Primary Industries, Water and Environment, Hobart. pp. 49–74. Hill, R.S., 1983. Nothofagus macrofossils from the tertiary of Tasmania. Alcheringa 7, Birks, H.H., 2013. Plant Macrofossils Introduction. In: Elias, S.A., Mock, C.J. (Eds.), Encyclope- 169–183. dia of Quaternary Science, second ed. Elsevier, Amsterdam, pp. 593–612. Hill, R.S., 2004. Origins of the southeastern Australian vegetation. Philos. Trans. R. Soc. Birks, H.J.B., Birks, H.H., 1980. Plant Macrofossils. In: Birks, H.J.B., Birks, H.H. (Eds.), Quaterna- Lond. B-Biol. Sci. 359, 1537–1549. ry Palaeoecology. Edward Arnold, London. Hill, R.S., Gibson, N., 1986. Distribution of potential macrofossils in Lake Dobson, Tasmania. Box, E.O., 2002. Vegetation analogs and differences in the northern and southern hemi- J. Ecol. 74, 373–384. spheres: a global comparison. Plant Ecol. 163, 139–154. Hill, R.S., Orchard, A.E., 1999. Composition and Endemism of Vascular Plants. In: Reid, J.B., Breda, N.J.J., 2003. Ground-based measurements of leaf area index: A review of methods, Hill, R.S., Brown, M.J., Hovenden, M.J. (Eds.), Vegetation of Tasmania. Melbourne, instruments and current controversies. J. Exp. Bot. 54, 2403–2417. Australia, Australian Biological Resource Study, pp. 89–124. Burnham, R.J., 1993. Reconstructing richness in the plant fossil record. PALAIOS 376–384. Huntley, B., 2001. Reconstructing past environments from the Quaternary palaeovegetation Carpenter, R.J., Horwitz, P., 1988. Leaf litter in two southern Tasmanian creeks and its record. Proc. R. Ir. Acad. 101B, 3–18. relevance to palaeobotany. Pap. Proc. R. Soc. Tasmania 122, 39–45. Jonckheere, I., Fleck, S., Nackaerts, K., Muys, B., Coppin, P., Weiss, M., Baret, F., 2004. Review of Chen, C., Qi, Z.C., Xu, X.H., Comes, H.P., Koch, M.A., Jin, X.J., Fu, C.X., Qiu, Y.X., 2014. methods for in situ leaf area index determination — part I. Theories, sensors and Understanding the formation of Mediterranean–African–Asian disjunctions: evidence hemispherical photography. Agric. For. Meteorol. 121, 19–35. for Miocene climate-driven vicariance and recent long-distance dispersal in the tertiary Jordan, G.J., 1997. Evidence of Pleistocene plant extinction and diversity from Regatta relict Smilax aspera (Smilacaceae). New Phytol. 204, 243–255. Point, western Tasmania, Australia. Bot. J. Linn. Soc. 123, 45–71. Christophel, D.C., 1984. Early tertiary Proteaceae: the first floral evidence for the Jordan, G.J., 2011. A critical framework for the assessment of biological palaeoproxies: Musgraveinae. Aust. J. Bot. 32, 177–186. predicting past climate and levels of atmospheric CO2 from fossil leaves. New Phytol. Christophel, D.C., Basinger, J.F., 1982. Earliest floral evidence for the Ebenaceae in 192, 29–44. Australia. Nature 296, 439–441. Jordan, G.J., Carpenter, R.J., Hill, R.S., 1991. Late Pleistocene vegetation and climate near Collinson, M., 1983. Accumulations of fruits and seeds in three small sedimentary environ- Melaleuca inlet, south-western Tasmania. Aust. J. Bot. 39, 315–333. ments in southern England and their palaeoecological implications. Ann. Bot. 52, Jordan, G.J., Macphail, M.K., Barnes, R., Hill, R.S., 1995. An early to middle Pleisto- 583–592. cene flora of subalpine affinities in lowland western Tasmania. Aust. J. Bot. 43, Collinson,M.E.,Manchester,S.R.,Wilde,V.,Hayes,P.,2010.Fruit and seed floras from 231–242. exceptionally preserved biotas in the European Paleogene. Bull. Geosci. 85, Kirkpatrick, J., 1982. Phytogeographical analysis of Tasmanian alpine floras. J. Biogeogr. 9, 155–162. 255–271. Cornelissen, J., Lavorel, S., Garnier, E., Diaz, S., Buchmann, N., Gurvich, D., Reich, P., Ter Macphail, M.K., 1979. Vegetation and climates in southern Tasmania since the last glaci- Steege, H., Morgan, H., Van Der Heijden, M., 2003. A handbook of protocols for ation. Quat. Res. 11, 306–341. standardised and easy measurement of plant functional traits worldwide. Aust. Mai, D.H., 1989. Development and regional differentiation of the European vegetation J. Bot. 51, 335–380. during the Tertiary. Plant Syst. Evol. 162, 79–91. Cornwell, W.K., Cornelissen, J.H.C., Amatangelo, K., Dorrepaal, E., Eviner, V.T., Godoy, O., Martin, R.E., 1999. Taphonomy: a Process Approach. Cambridge University Press. Hobbie, S.E., Hoorens, B., Kurokawa, H., Pérez Harguindeguy, N., Quested, H.M., McGlone, M.S., Turney, C.S., Wilmshurst, J.M., 2004. Late-glacial and Holocene vegetation 2008. Plant species traits are the predominant control on litter decomposition rates and climatic history of the Cass Basin, central South Island, New Zealand. Quat. Res. within biomes worldwide. Ecol. Lett. 11, 1065–1071. 62, 267–279. Cowling, R.M., Rundel, P.W., Lamont, B.B., Arroyo, M.K., Arianoutsou, M., 1996. Plant diversity McQueen, D.R., 1969. Macroscopic plant remains in recent lake sediments. Tuatara 17, in Mediterranean-climate regions. Trends Ecol. Evol. 11, 362–366. 13–19. Crowden, R.K., 1999. Alpine Vegetation. In: Reid, J.B., Hill, R.S., Brown, M.J. (Eds.), Vegetation Palamarev, E., 1989. Paleobotanical evidences of the Tertiary history and origin of the of Tasmania. Hobart, Australia, Tasmanian Government Printers, pp. 333–356. Mediterranean sclerophyll dendroflora. Plant Syst. Evol. 162, 93–107. G.A. Astorga et al. / Review of Palaeobotany and Palynology 233 (2016) 1–11 11

Read, J., 1999. Tasmanian Rainforest Ecology. In: Read, J.B., Hill, R.S., Brown, M.J., Spicer, R.A., Wolfe, J.A., 1987. Plant taphonomy of late Holocene deposits in trinity (Clair Hovenden, M.J. (Eds.), Vegetation of Tasmania. Australian Biological Resources Engle) Lake, northern California (USA). Paleobiology 13, 227–245. Study, pp. 160–197. Spicer, R.A., Wolfe, J.A., Nichols, D.J., 1987. Alaskan cretaceous-tertiary floras and Reich, P.B., Walters, M.B., Ellsworth, D.S., Vose, J.M., Volin, J.C., Gresham, C., Bowman, W.D., arctic origins. Paleobiology 13, 73–83. 1998. Relationships of leaf dark respiration to leaf nitrogen, specific leaf area and leaf Steart, D.C., Boon, P.I., Greenwood, D.R., Diamond, N.T., 2002. Transport of leaf litter in upland life-span: a test across biomes and functional groups. Oecologia 114, 471–482. streams of Eucalyptus and Nothofagus forests in southeastern Australia. Arch. Hydrobiol. Rich, F.J., 1989. A review of the taphonomy of plant remains in lacustrine sediments. Rev. 156, 43–61. Palaeobot. Palynol. 58, 33–46. Steart, D.C., Greenwood, D.R., Boon, P.I., 2005. Paleoecological implications of differential Roth, J., Dilcher, D., 1978. Some considerations in leaf size and leaf margin analysis of fossil biomass and litter production in canopy trees in Australian Nothofagus and Eucalyptus leaves. Cour. Forsch.-Inst. Senckenberg. 30, 165–171. forests. PALAIOS 20, 452–462. Rowell, M.V., Jordan, G.J., Barnes, R.W., 2001. An in situ,latePleistoceneMelaleuca fossil Steart, D.C., Greenwood, D.R., Boon, P.I., 2009. The chemical constraints upon leaf decay forest at coal head, western Tasmania, Australia. Aust. J. Bot. 49, 235–244. rates: taphonomic implications among leaf species in Australian terrestrial and Schimanski, L., Bergstrom, D., 1998. Nothofagus moorei (Nothofagaceae) communities: a aquatic environments. Rev. Palaeobot. Palynol. 157, 358–374. taphonomic interpretation. Aust. Biol. 11, 30–33. Stern, H., De Hoedt, G., Ernst, J., 2000. Objective classification of australian climates. Aust. Schneider, C.A., Rasband, W.S., Eliceiri, K.W., 2012. Nih image to imagej: 25 years of image Meteorol. Mag. 49, 87–96. analysis. Nat. Methods 9, 671–675. Stevens, P.F., Luteyn, J., Oliver, E.G.H., Bell, T.L., Brown, E.A., Crowden, R.K., George, A.S., Schnitzler, J., Barraclough, T.G., Boatwright, J.S., Goldblatt, P., Manning, J.C., Powell, M.P., 2004. Ericaceae. In: Kubitzki, K. (Ed.), The Families and Genera of Vascular Plants. Rebelo, T., Savolainen, V., 2011. Causes of plant diversification in the cape biodiversity Springer, Berlin, Germany, pp. 145–194. hotspot of South Africa. Syst. Biol. 60, 343–357. Weiller, C., 1996. Reassessment of Cyathodes (Epacridaceae).Aust.Syst.Bot.9,491–507. Sniderman, J.M.K., Jordan, G.J., Cowling, R.M., 2013. Fossil evidence for a hyperdiverse Weiller, C.M., 1999. Leptecophylla, a new for species formerly included in Cyathodes sclerophyll floraunderanon–Mediterranean-typeclimate.Proc.Natl.Acad.Sci.U.S.A. (Epacridaceae). Muelleria 12, 195–214. 110, 3423–3428. Wing, S.L., 1987. Eocene and Oligocene floras and vegetation of the Rocky Mountains. Sokal, R.R., Rohlf, F.J., 1995. Biometry.thirded.W.H.FreemanandCompany,San Ann. Mo. Bot. Gard. 748–784. Francisco. Wright, I.J., Cannon, K., 2001. Relationships between leaf lifespan and structural defences Spicer, R.A., 1981. The Sorting and Deposition of Allochthonous Plant Material in a Modern in a low-nutrient, sclerophyll flora. Funct. Ecol. 15, 351–359. Environment at Silwood Lake. Silwood Park, Berkshire, England. US Geological Survey, Wright, I.J., Reich, P.B., Cornelissen, J.H.C., Falster, D.S., Groom, P.K., Hikosaka, K., Lee, W., Professional Paper 1143. Lusk, C.H., Niinemets, Ã., Oleksyn, J., Osada, N., Poorter, H., Warton, D.I., Westoby, Spicer, R.A., 1989. The formation and interpretation of plant fossil assemblages. Adv. Bot. M., 2005. Modulation of leaf economic traits and trait relationships by climate. Res. 16, 95–191. Glob. Ecol. Biogeogr. 14, 411–421. Spicer, R.A., 1991. Plant Taphonomic Processes. In: Allison, P.A., Briggs, D.E.G.B. (Eds.), Ta- Xu, T., Hutchinson, M.F., 2013. New developments and applications in the Anuclim spatial phonomy: Releasing the Data Locked in the Fossil Record. Plenum Press, New York, climatic and bioclimatic modelling package. Environ. Model. Softw. 40, 267–279. pp. 71–113.

All in-text references underlined in blue are linked to publications on ResearchGate, letting you access and read them immediately.