Average chain length (ACL) of leaf wax components preserved in lacustrine sediments and soil profiles has been widely adopted as a proxy indicator for past changes in vegetation, environment and climate during the late Quaternary. The fundamental assumption is that woody plants produce leaf waxes with shorter ACL values than non-woody plants. However, there is a lack of systematic survey of modern plants to justify the assumption. Here, we investigated various types of plants at two lakes, Blood Pond in the northeastern USA and Lake Ranwu on the southeastern Tibetan Plateau, and found that the ACL values were not significantly different between woody and non-woody plants. We also compiled the ACL values of modern plants in the literatures and performed a meta-analysis to determine whether a significant difference exists between woody and non-woody plants at single sites. The results showed that the ACL values of plants at 19 out of 26 sites did not show a significant difference between the two major types of plants. This suggests that extreme caution should be taken in using ACL as proxy for past changes in vegetation, environment and climate.
The average chain length (ACL) of higher plants describes the average number
of carbon atoms based on the abundance of
Recently, a comprehensive review by Bush and McInerney (2013) summarized ACL
values for alkanes in modern plants from around the world, and found that ACL
was unable to distinguish graminoids (grasses) from woody plants. However,
single types of plants (either woody or non-woody) were present at ca.
57 % of the sites in the Bush and McInerney (2013) ranging from
tropical to temperate regions. It has been suggested that plants produce
longer-chain compounds in warmer climates (Poynter et al., 1989). Simoneit
et al. (1991) have analyzed continental aerosols from China and reported that
higher molecular weight
Here, we present ACL data for various types of modern plants at two lakes, Blood Pond in the northeastern USA and Lake Ranwu on the southeastern Tibetan Plateau, to determine whether or not ACL values differ between woody and non-woody plants. We then compiled published ACL data for 751 plants at 24 sites from around the world that contain both woody and non-woody plants to assess whether ACL could be a potential proxy for past changes in vegetation, environment and climate.
Lake Ranwu (29.48
Blood Pond (42.08
All leaf samples were freeze-dried and ultrasonically extracted (3
We compiled leaf wax ACL values for modern plants from the literatures following two criteria: (i) Investigations of natural plants or plants cultivated in an open field in a botanical garden. Greenhouse plants were excluded as they may suffer from too many human manipulations that would not reflect natural conditions. (ii) Investigations including both woody and non-woody plants at single sites. This would exclude the influences of various climate and environment factors on ACLs at different sites. Because not all original abundance data for plant alkanes or acids are available in literature, we cannot recalculate the ACL values for all plants, then we took the published ACL values in the publication. We intended to compare the ACL values between woody and non-woody plants at single sites. If the ACL values for plants were calculated from alkanes or acids of different carbon number ranges across different sites, the results would not affect the comparison in this study. The compiled ACL dataset included 751 plants at 24 sites, including 255 woody plants and 496 non-woody plants. Note that most of the ACL data were retrieved from Bush and McInerney (2013).
ACL values of leaf wax components are calculated from the following
equations (Poynter et al., 1989)
We divided the plants into three types according to growth habit,
including herbs, shrubs and trees. The average values for herbs,
shrubs and trees were 29.72, 29.36 and 26.93, respectively (Table 1,
Fig. 2a). The highest value was 32.10 for
The average acid ACL values for the various plant types at Blood Pond are
plotted in Fig. 2b. We follow the classification for plants by Hou
et al. (2007a, b) (Table 2). The plants included six types: ferns (2
samples), trees (16), shrubs (8), vines (3), herbs (9) and grasses (7). The
average ACL values were 27.56, 27.71, 27.54, 27.24, 28.07 and 27.59,
respectively. The highest was 29.40 (
We obtained ACL values for major types of plants in the literatures in order to examine the difference between woody and non-woody plants. Non-woody included, grasses and forbs, while woody plants included shrubs and trees. 751 plants from 24 sites from all continents except Antarctica met the criteria above (Table 3). Plant samples included 91 from Asia (Chikaraishi and Naraoka, 2003; Bi et al., 2005; Cui et al., 2008; Lei et al., 2010; Duan and Xu, 2012; Wilkie et al., 2013; Sarkar et al., 2014), 154 from Africa (Ali et al., 2005a; Vogts et al., 2009; Carr et al., 2014), 423 from Europe (Lütz and Gülz, 1985; Maffei, 1994, 1996a, b; Mayes et al., 1994; Ficken et al., 1998; Ali et al., 2005b; Nott et al., 2000), 9 from Oceania (Eglinton et al., 1962) and 74 from America (Salasoo, 1988; Nichols et al., 2006; Feakins and Sessions, 2010; Douglas et al., 2012) (see Supplement for details).
We performed meta-analysis on the 823 plants (529 herbs and 294 woody plants) from 26 sites (including the two case studies used in this study). The results showed that ACL values are statistically different between herbs and woody plants at 4 out of 26 sites (Table 3), including Sudan (Site J and O) (Ali et al., 2005a; Vogts et al., 2009), South China (Site L) (Bi et al., 2005) and Lake Qinghai, China (Site U) (Duan and Xu, 2012). As a meta-analysis cannot be performed on a dataset containing only one sample, three sites including Sudan (Site F) (Ficken et al., 1998), the USA (Site M) (Nichols et al., 2006) and China (Site S) (Lei et al., 2010) are different on the basis of visual observation. All other sites do not show a statistically significant difference in ACL values between woody and non-woody plants (Table 3).
It is surprising to observe that at the majority of the sites (
However, around 27 % of the surveyed sites (7 out of 26) do show a significant difference in ACL values between woody and non-woody plants by statistical analysis or visual observation. Examples include the Sudan (Ali et al., 2005a; Vogts et al., 2009), the USA (Nichols et al., 2006), the UK (Ficken et al., 1998) and China (Duan and Xu, 2012; Lei et al., 2010; Bi et al., 2005) (F, J, L, M, O, S, U in Fig. 3 and Table 3). It unclear why the plants at the seven sites show a difference in ACL values between plants, which probably implies that ACL could be used to differentiate plant types and vegetation dynamics in a specific area. However, ACL values behave differently for plants collected in botanical gardens. At a botanical garden in Italy (Site D), ACL is not statistically different between woody plants and non-woody plants (Maffei, 1994, 1996a, b), while ACL values are statistically different in the South China Botanical Garden (Site L) (Bi et al., 2005). This may be the result of a significant difference in the irrigation regime in botanical gardens. In addition, visual observation revealed that the ACL values of non-woody plants are lower than those of woody plants in the UK (Site F) (Ficken et al., 1998), in contrast to the assumption that woody plants have lower ACL values than non-woody plants.
ACL values of terrestrial plant leaf waxes could be affected by many
environmental and plant physiological factors that would make them
indistinguishable between woody and non-woody plants. Temperature and
precipitation and their combinations have been found to affect the
lipid distributions of terrestrial plants. Plants tend to
biosynthesize longer chain lipid compounds for waxy coatings in warm
regions and shorter chain lipids in cool regions (Castañeda
et al., 2009). Plants in warmer climates have been suggested to
produce longer-chain compounds (Poynter et al., 1989). Simoneit
et al. (1991) found that the plants in warmer climate of southern
China produced higher molecular weight
Regional precipitation could also influence the chain length
distribution of leaf wax lipids. Under water stress, plants tend to
synthesize longer carbon chain alkanes in order to provide a more
efficient waxy coating (Calvo et al., 2004). For instance, atmospheric
dust samples collected in transects along the west African coast
reveal changes in the chain length distributions of
For those plants with the same growth habit, the lipid distribution
may be distinct. For example, coniferous trees tend to produce longer
carbon chain lipids than broad-leaf plants (Zhang et al., 2008). The
ACL could vary significantly even in leaves from individual plants of
the same genus. Hoffmann et al. (2013) found that ACL varied from
26.87 to 28.99 for
The average chain length of plant lipids that are preserved in various sediments has been widely used as a proxy indicator for past changes in vegetation, environment and climate. However, the fundamental assumption of using ACL as a proxy indicator seems problematic based on the meta-analysis of two case studies and a data set compiled from around the world. About 73 % of the surveyed sites (19 out of 26 sites) and over 90 % of plants (741 out of 823 plants) do not show a statistical difference in ACL values between woody plants and non-woody plants. Nevertheless, about 27 % of the surveyed sites and less than 10 % of plants do show a difference on a global basis. The results suggest that considerable caution is necessary in using ACL values as a proxy indicator for vegetation dynamics, and for interpreting ACL variation in terms of past changes in environment and climate.
J. Hou conceived and designed the experiments. M. Wang and W. Zhang performed the experiments. J. Hou, M. Wang analyzed the data. J. Hou, M. Wang wrote the manuscript.
This work was supported by grants from the Chinese Academy of Sciences (KZCX2-YW-JC106 to J. Hou) and National Natural Science Foundation of China (41072120 to J. Hou).
ACL values for modern plants at Lake Ranwu on the southeastern Tibetan Plateau, China.
ACL values for modern plants at Blood Pond in the northeastern United States.
Continued.
Locations of Blood Pond in the northeastern United States and Lake Ranwu on the southeastern Tibetan Plateau, China.
Distribution of ACL values for modern plants at Lake
Ranwu
Compiled dataset of modern plants from around the world (including Lake Ranwu and Blood Pond in this study). Letters correspond to the sites in Table 3, and the numbers of samples at each site are listed in parentheses. Pie charts show the percentage of woody and non-woody plants at each site. Stars denote the fact that there is a statistically significant difference in ACL values between the two types of plants.
Distribution of ACL values for modern plants at 24 sites around the world. Solid diamonds represent non-woody plants (including graminoids/grasses and forbs), and open diamonds represent woody plants (including shrubs and trees). Letters correspond to the sites in Table 3 and Fig. 3.