Next Article in Journal
Hardwoods: Anatomy and Functionality of Their Elements—A Short Review
Previous Article in Journal
Application of Random Forest Method Based on Sensitivity Parameter Analysis in Height Inversion in Changbai Mountain Forest Area
 
 
Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
Journals
Forests
Volume 15
Issue 7
10.3390/f15071163
forests-logo
Submit to this Journal Review for this Journal Propose a Special Issue
► Article Menu

Article Menu

share Share announcement Help format_quote Cite question_answer Discuss in SciProfiles thumb_up
...
Endorse textsms
...
Comment
first_page
settings
Order Article Reprints
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

Modern Lacustrine Phytoliths and their Relationships with Vegetation and Climate in Western Yunnan, SW China

by
Yanyan Xu
1,
Caiming Shen
1,*,
Min Wang
1,*,
Hongwei Meng
1,
Qifa Sun
2,
Linpei Huang
1 and
Huiling Sun
1
1
Yunnan Key Laboratory of Plateau Geographical Processes and Environmental Changes, Faculty of Geography, Yunnan Normal University, Kunming 650500, China
2
College of Resources, Environment and Chemistry, Chuxiong Normal University, Chuxiong 675000, China
*
Authors to whom correspondence should be addressed.
Forests 2024, 15(7), 1163; https://doi.org/10.3390/f15071163
Submission received: 22 May 2024 / Revised: 27 June 2024 / Accepted: 1 July 2024 / Published: 4 July 2024
(This article belongs to the Section Forest Ecology and Management)
Browse Figures
Versions Notes

Abstract

:
As a plant kingdom and a biodiversity hotspot, Yunnan is a key region for our understanding of modern and past global changes in biodiversity and environment. As proxies of vegetation and climate, phytoliths have become increasingly important in ecological and paleoecological studies. In this study, phytolith analysis was carried out on samples of surface sediments from 70 lakes in western Yunnan, southwest China. These lakes are surrounded by modern vegetation types including broadleaved and coniferous forests, scrubs, grasslands, meadows, and alpine vegetation. The results of this study show that modern lacustrine phytoliths in western Yunnan are dominated by herbaceous phytoliths, among which Poaceae types are the most abundant. The 70 phytolith samples used can be divided into 4 groups, reflecting the major vegetation types from which samples were collected. The principal component analysis (PCA) and redundancy analysis (RDA) of the phytolith and climatic data of the 70 lacustrine phytolith samples showed that temperature and precipitation are the climatic parameters controlling the spatial distribution of phytolith assemblages in western Yunnan. Phytolith–MAT (mean annual temperature) and phytolith–MAP (mean annual precipitation) transfer functions were developed using weighted averaging partial least squares (WA-PLS), and both the MAT and MAP functions showed good performances (MAT: R2 = 0.67, RMSEP = 0.96 °C, MAP: R2 = 0.64, RMSEP = 140.4 mm). Our results also reveal that phytolith analysis is a useful technique offering reliable vegetation interpretation and climate reconstruction; thus, this study provides a basis for the vegetational and climatic interpretation of fossil lacustrine phytolith records in western Yunnan.

1. Introduction

Lakes are “sinks” of microscopic particles produced mainly by surrounding vegetation [1]. In addition to the substances that deposit in situ, the sediments brought by runoffs and rivers into lakes include local and regional substances [2]. Some substances are even transported to lakes by wind from relatively open and burning environments [3]. Phytolith analysis, at its early stage, was mainly used in archaeological and paleoecological studies of archaeological layers and loess as vegetation and climate proxies [4,5], due to their poor preservation of plant macrofossils and pollen but abundance of phytoliths. Phytoliths, composed of biogenic silica that are precipitated in or between cells of higher plants [6], have the features of high yield, wide distribution, and resistance to high temperatures, acid, and alkali [7]. Today, phytolith analysis is also applied within paleoecological studies of lacustrine sediments [8,9,10].
As a biodiversity hotspot and one of the most biodiverse regions in the world [11,12], western Yunnan—in the southeastern margin of the Tibetan Plateau—is a typical low-latitude mountainous region. Its terrain is complex, and mountains and deep river valleys occur alternately, along with an abundance of lakes [13]. Furthermore, it is sensitive to climate change and human activities [14]. In the past decades, many studies have been carried out to reveal regional vegetation and climate change in Yunnan, and most fossil records used in these studies were mainly taken from lakes [15,16,17]. Therefore, phytolith analysis of lacustrine sediments can offer local to regional information about the vegetation and climate surrounding lakes. In order to interpret fossil phytolith data of lacustrine sediments in terms of vegetation and climate, it is essential to study the associations of modern lacustrine phytolith assemblages with vegetation and climate. At present, phytolith analysis of modern lacustrine sediments has provided some information on the understanding of the relationships between phytolith assemblages and vegetation and climate in some regions [18,19,20]. However, such studies are very limited; more studies, particularly involving modern lacustrine phytolith assemblages and their relationships with vegetation, climate, and environment in different regions, are essential in order to offer modern analogs of such associations for the reconstructions of paleovegetation, paleoclimate, and paleoenvironment.
This study presents modern lacustrine phytolith assemblages from 70 lakes in western Yunnan, to determine the characteristics of modern lacustrine phytolith assemblages and quantitatively elucidate the associations of modern lacustrine phytoliths with the vegetation and climate in this region. The major objectives of this study are as follows: (1) to determine the major features of 70 phytolith assemblages of lake surface sediments; (2) to determine the associations of modern lacustrine phytolith assemblages with the modern vegetation and climate in the lake catchments; (3) to develop a set of phytolith–climate transfer functions for future quantitative reconstructions of paleoclimate in western Yunnan.

2. Study Region

2.1. Topography

Western Yunnan [21°10′–29°15′ N, 97°31′–102°19′ E, 204–6457 m a.s.l. (above sea level)] (Figure 1a) is a typical low-latitude mountainous region in southwest China. Its overall topographic characteristics are high altitudes in the north and low in the south; large undulation, with alternating mountains and rivers; and complex and diverse landform types. Its north is the southern extension of the Tibetan Plateau, with altitudes of about 3000~4000 m a.s.l. and a typical alpine-canyon landform. The Hengduan Mountains extend from north to south, and are separated by the rivers Nujiang, Lancangjiang, and Jinshajiang. The terrain with middle mountains and wide valleys is relatively fragmented in its central part. The terrain of southwest Yunnan is gradually flat, and the open river valley areas develop under the control of the “broom-shaped” mountain system [21], coinciding with the mountain range extensions in southeast Yunnan [22].

2.2. Climate

The climate in western Yunnan has obvious low latitude and high plateau characteristics. In the winter half-year, the southward dry and cold air flow is blocked by the Tibetan Plateau and the Yunnan Plateau, resulting in a warm climate. In the summer half-year, it is controlled by the South Asian summer monsoon (SASM) and the significant accompanying rainfall. In addition, western Yunnan has obvious mountain climate characteristics [23], and its temperature and rainfall vary greatly at different altitudes and slopes. Therefore, the study region has the characteristics of low-latitude, monsoon, and mountain climates [24]. In general, the three-dimensional climate in western Yunnan has a large daily temperature difference, a small annual temperature difference, and distinct dry and wet seasons. Observational data from 124 meteorological stations in Yunnan across 30 years (1982–2011) (Figure 1b,c) have shown that the mean annual temperature (MAT) in this region is 10.4–20.2 °C, and the mean annual precipitation (MAP) is 716–1691 mm, which is mainly concentrated in the rainy season (May–October).

2.3. Vegetation

The uniqueness of the climate and the complexity of the geographical environment result in diverse types and characteristics of regional vegetation [25]. Vegetation distribution, in turn, also reflects the comprehensive natural environmental conditions [26]. Western Yunnan has diverse environments and vegetation types [27], ranging from the tropical rainforest and monsoon forest in its south, to the subtropical evergreen broadleaved forest in its center, and to the high-cold vegetation of the Tibetan Plateau in its north, following the south–north gradients of temperature and precipitation. In the lower altitude valleys in its south, the vegetation is predominantly adapted to dry and warm conditions. Evergreen broadleaved forest can be found in intermediate altitude areas. Yunnan pine forest becomes more prominent in its north, while more scrub grows as the altitude increases. At higher altitudes, there are more frequent occurrences of dense coniferous forest, meadows, and alpine vegetation, which are habituated to the slightly drier and colder climate of mountains (Figure 2).
Our study region encompasses four vegetation zones [26,28]: zones A, B, C, and D, which are dominated by northern tropical seasonal rainforests and semi-evergreen monsoon forests; southern subtropical monsoonal evergreen broadleaved forests; northern subtropical semi-humid evergreen broadleaved forests; and cold–temperate coniferous forests and meadows (Figure 2), respectively. These four vegetation zones also belong to the three vegetation districts in China, i.e., the districts of tropical monsoon forest and rainforest; of subtropical evergreen broadleaved forest; and of the high-cold vegetation in the Tibetan Plateau. Different dominant species occur in different vegetation zones (Table 1).

3. Materials and Methods

3.1. Sampling and Sample Processing

A total of 70 lake surface samples were taken, using a gravity sampler, from 70 lakes [including natural or artificial lakes (reservoirs)] during field trips in 2015 (14), 2017 (22), and 2018 (34). They were sampled from the uppermost ca. 5 cm sediments at the center and/or deepest part of each lake, which accumulated ca. over less than 50 years, so the phytolith assemblage of 1 sample is a mixture of phytoliths over the recent decades, i.e., the average state of recent decades. The locations and water depths of the sampling lakes were measured using GPS and sounding apparatus. The vegetation in the catchments of the 70 lakes cover all the main vegetation zones of western Yunnan. The lake water’s history, area, and storage were also recorded in detail while collecting the surface sediments of the lakes.
The extraction and identification of phytoliths were completed at the Key Laboratory of Plateau Geographical Processes and Environmental Changes, Yunnan Normal University. For each sample, 0.2 g of dried lake sediments were moved into a 50 mL centrifuge tube. Samples were then processed through treatment with 30% H2O2 and 10% HCl using the conventional wet oxidation method [29], followed by heavy liquid flotation using ZnBr2 of 2.35 g/cm3. Extracted phytoliths were permanently mounted on glass microscopic slides with balsam, and counted using a microscope at 400× magnification.

3.2. Environmental Data

According to the 30-a (1982–2011) observational data from 124 meteorological stations in Yunnan, the values of climatic parameters for each lake were obtained using ArcGIS. The potential climatic parameters include MAT, MAP, average temperature, and precipitation in January (Tjan and Pjan), July (Tjuly and Pjuly), the rainy season (May–October) (Trainy and Prainy), and the dry season (November–April) (Tdry and Pdry). Additionally, the altitude of the lakes was also chosen as a climatic parameter.

3.3. Statistical Analysis

Canonical ordination techniques, using the statistical program R software (version 4.3.0) (Vienna, Austria), were used to identify covarying patterns between phytolith assemblages and the 11 climatic parameters. Firstly, detrended correspondence analysis (DCA) was conducted in order to determine which model (linear or unimodal) was most appropriate, with gradient length (range of variation) as the criterion [30]. It was thereby shown that the linear models were preferable for our phytolith dataset, given that their gradient lengths of standard deviation units are smaller than the two revealed by DCA [31]. PCA is used to reduce the dimensionalities of high-dimensional datasets in a range of research areas [32]. PCA is similar to a clustering method, as it finds patterns without reference to prior knowledge about whether the samples come from different groups or have phenotypic differences. Redundancy analysis (RDA) was then used to study the variations in the biological assemblages that can be explained by a particular set of environmental variables. It can visualize complex phytolith data and their relationships to the environmental variables (e.g., MAT, MAP). The resulting ordination diagrams of the RDA show the main patterns of variation in the compositions of the phytolith assemblages, as accounted for by the environmental variables and, in an approximate way, the distributions of the phytoliths along each environmental variable. In order to remove the effects of high collinearity among the environmental variables in the analyses, we examined the variance inflation factors (VIFs) for each environmental variable. If the VIF value of a variable was larger than 20, the variable was assumed to be almost perfectly correlated with the other variables and captured little variance [33]. After the first screening of the 11 environmental variables, the VIF values of 4 variables were lower than 20, suggesting that 4 environmental variables had a unique influence on the distribution of phytoliths.
To generate phytolith–climate transfer functions, six models were tested for the dataset using C2 (version 1.5) (Newcastle, United Kingdom) [34]. The six models include weighted averaging (WA); weighted averaging partial least squares (WA-PLS); Imbrie and Kipp factor analysis regression (IKR); modern analogue technique (MAT); maximum likelihood (ML); and locally weighted weighted-averaging (LWWA). In contrast to simple WA, WA-PLS uses additional components that utilize the residual structure in the biological data to update the species optima and reduce the error and the patterns in the bias of the final calibration function [35,36,37]. We selected the final WA-PLS model (appropriate number of components) that gave the lowest root mean squared error of prediction (RMSEP) and the largest adjusted coefficients of determination, R2_Jack, as assessed by the leave-one-out cross-validation method [38].

4. Results

4.1. Phytolith Types

A minimum of 500 phytoliths were counted for each sample. In this study, phytoliths were classified according to the system proposed by Wang and Lu [14], and described according to international phytolith nomenclature (ICPN1.0) [29,39] (Figure 3). In this study, the phytoliths were categorized into three plant groups, i.e., herbs, woods, and ferns.
(1)
Herb silica short cells are produced by Poaceae [40]. Poaceae phytoliths were further subdivided into six subfamilies [13]: Dumbbell (Figure 3f,g) and Cross (Figure 3e) types, which are mainly produced by Panicoideae—C4 grasses adapted to warm and humid climates; Rondel (Figure 3i) and Trapezoid (Wavy Trapezoid and Wavy Narrow Trapezoid) (Figure 3s,t) types, which are produced by Pooideae [41] and are good indicators of a cold and dry climate; the Short Saddle (Figure 3a,b) type, which occurs predominantly in Eragrostoideae, which indicates a warm and dry climate; and the Long Saddle (Figure 3c,d) and Fan Bamb types, which are produced by the Bambusoideae, Saddle, and Fan Reed types. These are produced by Arundinoideae, as well as the Rice Bulliform, Double Peaked, and Parallel Bilobate types produced by Oryzoideae, all of which are good indicators of a warm and wet climate. Other grass phytoliths include the Square (Figure 3q), Rectangle (Figure 3r), Sinuate Elongate (Figure 3x,y), Smooth Elongate (Figure 3u–w), and Point (Figure 3m,n) types. Dumbbell, Square, and Rectangle forms are mainly deposited in bulliform cells of the subfamilies Oryzoideae, Arundinoideae, and Panicoideae. Bulliform cells can be used as an index of warm and wet climates [42], because the high production of bulliform cells is related to higher water availability [43]. Herb phytoliths also contain Sedge types.
(2)
Woods include broadleaved types, Palmae types, and Gymnosperm. Phytoliths of broadleaved types (Figure 3ab–ae) are produced by many tropical and subtropical broadleaved trees. They are composed of Net Spindle, Spiral Spindle, Abbreviated Stellate, and Polygonal Plate types (Figure 3ab,ac). According to studies of phytolith assemblages in Chinese surface soil [44,45], Dumbbell, Square, Rectangle, Long Saddle, Fan Bamb and broadleaved types can be used as indicators of wet and warm climates. Palmae types, adapted to wet and warm climates, are composed of globular echinate. Tabular striate (Figure 3af) is produced by Gymnosperm.
(3)
Ferns produce geniculate phytolith (Figure 3ag,ah), reflecting a warm and wet climate.
Forests 15 01163 g003
Figure 3. Major phytolith morphotypes, identified for phytoliths, form surface sediments of 70 lakes in western Yunnan. (a,b) Short Saddle; (c,d) Long Saddle; (e) Cross; (f,g) Dumbbell; (h) Polylobate; (i) Rondel; (j–l) Tower; (m,n) Point; (o,p) Bulliform; (q) Square; (r) Rectangle; (s,t) Wavy Trapezoid; (u–w) Smooth Elongate; (x,y) Sinuate Elongate; (z) Board Elongate; (ab,ac) Polygonal Plate, broad-leaved types; (ad) Globular Smooth, broadleaved types; (ae) “Y” type, broadleaved types; (af) Tabular striate, Gymnosperm types; (ag,ah) Pteridophyte types. Black bar represents 10 µm.
Figure 3. Major phytolith morphotypes, identified for phytoliths, form surface sediments of 70 lakes in western Yunnan. (a,b) Short Saddle; (c,d) Long Saddle; (e) Cross; (f,g) Dumbbell; (h) Polylobate; (i) Rondel; (j–l) Tower; (m,n) Point; (o,p) Bulliform; (q) Square; (r) Rectangle; (s,t) Wavy Trapezoid; (u–w) Smooth Elongate; (x,y) Sinuate Elongate; (z) Board Elongate; (ab,ac) Polygonal Plate, broad-leaved types; (ad) Globular Smooth, broadleaved types; (ae) “Y” type, broadleaved types; (af) Tabular striate, Gymnosperm types; (ag,ah) Pteridophyte types. Black bar represents 10 µm.
Forests 15 01163 g003

4.2. Phytolith Assemblages

The phytolith spectra from modern lacustrine sediments are dominated by grass phytoliths, mainly including Poaceae phytoliths such as Long Saddle, Short Saddle, Dumbbell, Smooth Elongate, and Rondel types. The 70 phytolith spectra are divided into four groups based on the results of the CONISS analysis of phytolith percentage data (Figure 4):
(1)
Group I: It consists of 24 phytolith spectra, characterized by Long Saddle (30.06%, the average of group phytolith spectra, the same below), Dumbbell (20.98%), Elongate (17.40%), Palmae (3.60%), and Tabular Striate (2.39%) phytoliths. The Long Saddle type had the highest percentage, followed by Dumbbell, then Elongate. This group is characterized by the highest herbaceous phytolith percentage (92.0%). The high percentages of grasses such as the Long Saddle and Dumbbell phytoliths suggest the occurrence of large amounts of grass and bamboo.
(2)
Group II: It is composed of 11 phytolith spectra, characterized by the Dumbbell (21.99%), Long Saddle (21.88%), Tabular Striate (3.44%) and Palmae types (2.16%). The occurrence of Palmae type phytoliths implies high temperature conditions [46]. The higher percentages of Dumbbell and Long Saddle in this group compared to group I suggest a greater quantity of grasses and bamboos. Higher grasses may be resultant from frequent and repeated human cultivation. The occurrence of Sedge (0.78%) implies humid environments. This group is characterized by the lowest amount of broadleaved phytoliths (4.87%).
(3)
Group III: This group includes 6 phytolith spectra. It is characterized by Short Saddle (18.75%), Rondel (12.07%), Dumbbell (10.85%), Tabular Striate (3.28%), Globular Smooth (2.57%), and Palmae (1.10%) phytoliths. The low percentage of Dumbbell and high percentage of Short Saddle likely reflect the warm and dry climate. The percentages of many types in this group, such as the Rondel (12.07%) and Tower (13.85%) phytoliths, indicate the predominance of cold climate-adaptable plants.
(4)
Group IV: It consists of 29 phytolith spectra, characterized by the highest amount of broadleaved type (5.74%) and the lowest amount of herbaceous (87.74%) phytoliths. Elongate (17.48%), Long Saddle (16.30%), Dumbbell (12.28%), Rondel (11.61%), Tabular Striate (4.44%), Palmae (2.98%), and Tabular Elongate Cavate (1.35%) phytoliths are abundant. The highest numbers of broadleaved types are found in this group, implying the predominance of evergreen broadleaved plants. Gymnosperm types (6.53%) are high, while Stipa type (0.44%) and Long Saddle (16.3%) phytoliths are low, indicating flourishing Gymnosperm forest with understory Pooideae. The phytolith spectra of this group contain a high quantity of unknown types, indicating that many varieties from this wide range of phytolith morphotypes are yet to be investigated.
Forests 15 01163 g004
Figure 4. Phytolith percentage diagram of major phytolith types forming surface sediments of 70 lakes in western Yunnan.
Figure 4. Phytolith percentage diagram of major phytolith types forming surface sediments of 70 lakes in western Yunnan.
Forests 15 01163 g004

4.3. Ordination Analysis

In order to clarify the similarities in phytolith assemblages, as well as the main patterns of phytolith variation, the percentages of major phytolith types that have average phytolith percentages of more than 1% for all samples were first analyzed by PCA. The major phytolith types include Elongate, Cross, Point, Square, Rectangle, broadleaf type, Dumbbell, Gymnosperm type, Long Saddle, Rondel, Short Saddle, Polylobate, Sedge types, Wavy Trapezoid, and Bullform phytoliths. The PCA results show that the first two components explain 32% and 15% of the total variance of this phytolith dataset (Figure 5). As shown in Figure 5, the Gymnosperm, Wavy Trapezoid, Short Saddle, Rondel, and Point types have positive loadings on the PC1, whereas Dumbbell and Long Saddle have negative loadings. PC2 separates Square, Rectangle, and Bulliform from Polylobate and Cross phytoliths, implying changes in the minor components of the above phytolith types. PC1 also clearly distinguishes samples of groups I, III, and IV, and PC2 separates samples of group II from the other groups. Groups I and II reflect the vegetation indicating warm and wet climatic conditions, whereas groups III and IV reflect the vegetation indicating cold and dry climatic conditions.
In order to clarify the relationships between phytolith assemblages and climatic parameters in western Yunnan, an RDA was conducted on a dataset consisting of climatic parameters and the percentage data of the major phytolith types. Monte Carlo permutation tests (999 unrestricted permutations) of the first two axes indicated that both axes were statistically significant (p < 0.01). Forward selection and unrestricted Monte Carlo permutation tests indicated that four climatic parameters (MAP, MAT, Pjan, and Altitude) are statistically significant in relation to the variance in the phytolith data. Permutation tests were conducted on the RDA model to assess the statistical significance of the climatic constraining parameters as predictors of the variation in phytolith assemblages [47,48]. The eigenvalues of the first two RDA axes are λ1 = 0.245 and λ2 = 0.015, and the species–environment correlations for the first axis are high (0.877), indicating strong relationships between the distribution of the 15 phytolith types and the four climatic parameters, especially MAT and MAP (Table 2).
In the RDA species—and sample—environment biplot (Figure 6), the lengths of the arrows for the climatic parameters roughly correspond to their relative importance in phytolith variance, and their orientations show their approximate correlations to ordination axes and the other climatic parameters. MAP is closely related to MAT, since there is only a small angle between the arrows of MAP and MAT. Altitude is negatively correlated with MAP, MAT, and Pjan. Phytolith types preferring high MAP and MAT are generally positioned to the lower left of the RDA biplot (e.g., Dumbbell, Bulliform, and Cross phytoliths). They are in the highest abundance in lakes over the south and southwest of Yunnan, where seasonal rainforest, semi-evergreen monsoon forest, and monsoonal evergreen broadleaved forest dominate. The phytolith types situated in the right of the biplot are Rondel, Point, Wavy Trapezoid, and Gymnosperm phytoliths, which have the highest abundance in the semi-humid evergreen broadleaved forest, cold-temperate coniferous forest, and meadow of northwestern Yunnan.
The RDA results clearly indicate that MAP and MAT are the most significant climatic parameters controlling our modern lacustrine phytolith data. Therefore, the relationships of modern phytolith distribution with MAP and MAT are sufficiently robust enough to develop regional phytolith–MAT/MAP transfer functions to assist with quantitative reconstructions of the paleoclimate using phytolith records from lakes in western Yunnan.

4.4. Weighted Averaging-Partial Least Square Regression and Calibration

In order to select the best calibration model for our phytolith and MAT/MAP dataset, six models were used with this dataset. The results (Table 3) show that the WA-PLS model performed the best for MAT and MAP.
Finally, the WA-PLS calibration model was selected for the establishment of phytolith–MAT/MAP transfer functions. In this procedure, 13 samples, i.e., DX-1, DX-2, DX-3, DX-4, DX-8, DX-9, DX-20, DX-23, DX-25, DX-26, DX-28, DX-29, and DX-34, having very high absolute residuals of MAT, were deleted as outliers; 15 samples, i.e., DX-4, DX-6, DX-7, DX-9, DX-20, DX-21, DX-22, DX-23, DX-24, DX-25, DX-26, DX-28, DX-29, DX-51, and DX-68, having very high absolute residuals of MAP, were deleted again as outliers; the final training dataset included 50 samples and 15 phytolith types. The performance of the phytolith–MAT/MAP transfer function developed from this final training dataset is summarized in Table 4 and Figure 7. The low RMSEP (0.96 °C and 140.4 mm for MAT and MAP) and high adjusted coefficients of determination R2-Jack (0.67 and 0.64 for MAT and MAP) suggest that this training dataset and WA-PLS model can be used to develop phytolith–MAT/MAP transfer functions.

5. Discussion

5.1. Phytolith Taphonomy

In order to accurately evaluate the phytolith assemblages, their taphonomy must be taken into consideration. Lake sediment records are affected by various taphonomic processes, which involve the production, transport, deposition, and conservation of proxies in sediments [49].
The differential production of phytolith morphotypes can influence phytolith assemblages. Different taxa produce different quantities of phytoliths [29,50]. The large forest around the lakes did not mask the grass signal, probably due to the high phytolith production of Poaceae [29] and low production of forest trees. This clearly has an important implication for the interpretation of phytolith assemblages from forests.
Transport and deposition of phytoliths could influence lacustrine phytolith assemblages as well. Understanding the major transportation mode of phytoliths in lakes is a crucial step for interpreting modern and fossil phytolith assemblages from lakes. Phytoliths generally do not travel too far compared to other microscopic materials because they are relatively heavy [49]. However, they may still be transported long distances by wind. In dry areas with rare plants and frequent high winds, wind transport is common [50]. Additionally, phytoliths are already found in atmospheric dust [51,52]. This mode of transport, however, is uncommon in wooded areas [53]. Forest fires can also result in wind-borne transport of phytoliths [54].
Conservation bias may also result in some degree of influence on phytolith assemblages. Misidentification may be resultant from the mixture of phytoliths with non-phytolith materials, such as quartz, tephra, volcanic silica, spores, diatoms, and other microscleres [55]. Volcanic silica is similar to blocky, irregular, and tabular phytoliths; it was difficult to separate them from the phytoliths. Modern phytoliths were found to be silicified lightly and less stable [56,57,58,59] due to the recycling of the silica [60,61]. Fragile phytoliths might be dissolved under the influences of silica saturation of the sediments, the pH values of the environment, and mineral impurities [62,63,64,65]. It should be pointed out that the variability in phytolith preservation across different environmental conditions might impact phytolith assemblages, since the production, transport, deposition, and preservation of phytoliths in sediments is still a challenge in the vegetational and climatic interpretation of phytoliths [7,8].

5.2. The Relationship between Phytolith Assemblages and Vegetation

Differentiating environments is a fundamental goal of phytolith studies, in spite of the problems with phytolith taphonomy. Certainly, identifying environments by phytolith assemblages is not easy in subtropical and tropical regions, due to the overlapping production of phytolith morphotypes in various plant groups. Even so, differences of phytolith morphotypes are valuable tools in explaining vegetation change [46,66,67,68].
This study set out to test whether regional vegetation in the study region could be distinguished by modern lacustrine phytolith assemblages. Some morphotypes are indicative of particular vegetation types. The majority of grasses growing in wet and warm regions of southwest China are C4 Panicoideae and Chloridoideae [69]. The Saddle type is most strongly represented in the samples, which fits the association between Chloridoideae grasses and Saddle phytoliths. Modern lacustrine phytolith assemblages in this study are dominated by high percentages of Long Saddle and Dumbbell, in which the former is locally diagnostic for Bambusoideae grasses, and the latter is most likely derived from Panicoideae grasses. Dumbbell and Long Saddle morphotypes were most common in the samples from vegetation zones A and B. It is not unexpected that the high temperature and precipitation along with warm summers and mild winters in these two vegetation zones result in extremely dense understory vegetation in forests. High percentages of Trapezoid, Rondel, and Point types may represent woodland grasses, which are common in vegetation zones C and D. It is similar to situations in earlier studies in other regions where phytoliths of temperate Pooideae dominated [70,71,72]. Gymnosperm types have the highest proportion in vegetation zones C and D, consistent with regional vegetation in more coniferous forests.
The results of the PCA and RDA further support the relationship of phytoliths with vegetation deduced from the examination of the proportions of phytoliths. Samples of groups I and II were separated from groups III and IV largely by PC1 and RDA1. Samples of phytolith groups I and II were at the negative end, whereas samples of groups III and IV were at the positive end of phytolith PC1 and RDA1. Figure 8 displays how phytolith groups appear spatially, which appears similar to the spatial pattern of vegetation zones. Samples of phytolith groups I and II mainly occurred within vegetation zones A and B, and groups III and IV within vegetation zones C and D. This fact suggests that lacustrine phytolith assemblages can be used to explain regional vegetation. However, the difference in phytolith assemblages between vegetation zones A and B is not significant, given the minor difference in vegetation types between them, especially the understory vegetation which is well represented by phytoliths. Certainly, more studies on phytolith responses to vegetation are needed.
Most of the 70 lakes selected are located in the regions of western Yunnan that have Human Influence Index (HII) values less than 22 [73], where vegetation is dominated by native species [74], so human activities such as agriculture and deforestation have little potential influence on modern phytolith assemblages. This study will be helpful for providing a better understanding of where lacustrine phytoliths come from. The identification of phytolith types can provide ecological information on ecosystem differentiation and archeological information on human activities [75,76,77]. Some studies seem to support our conclusions. Blinnikov [78] and Morris et al. [79] distinguished surface grassland, shrublands, and forests of temperate regions by this method. Whilst in these lakes phytolith analysis alone was not able to differentiate all ecosystems, it has added taxonomic value.

5.3. The Relationship between Phytolith Assemblages and Climate

In general, the vegetation types from which samples of phytolith groups were collected grow within specific ranges of precipitation and temperature. Since vegetation types can be well represented by phytoliths, as discussed above, the associations between phytoliths and climatic parameters can be determined by RDA. RDA separates phytolith groups I and II in warm–wet climatic conditions, and groups III and IV in cool–dry climatic conditions. In the species–samples–environment biplot of the RDA, the lengths of the arrows for the environmental variables roughly correspond to their relative importance in phytolith variance, and their orientations show their approximate correlations to ordination axes and other environmental variables. MAT and MAP are closely related since there is only a small angle between their arrows. Phytolith types preferring high MAP and MAT are generally positioned to the left of the RDA biplot (e.g., Dumbbell and Bulliform phytoliths). They are the most abundant at lakes in southwest Yunnan, where vegetation is dominated by tropical seasonal rainforest, semi-evergreen monsoon forest, and subtropical monsoonal evergreen broadleaved forest. The phytolith types situated in the right of the biplot are Point, Rondel, Wavy Trapezoid, and Gymnosperm types, which have the highest abundance in the subtropical semi-humid evergreen broadleaved forest, the cold–temperate coniferous forest, and the meadow in northwest Yunnan. This further confirms the association between phytoliths and vegetation as well as climate.
Even though phytoliths are not identified at species level, they still enable the inference of climatic conditions. Phytoliths are good tools for interpreting climatic regimes, quantitatively estimating MAP and MAT [44,45], and determining wet and dry periods [80]. As shown by Figure 1b,c, climatic conditions in western Yunnan have only two modes, i.e., “warm-wet” and “cold-dry”. Phytolith groups I and II are characterized by high MAT and MAP. Their sampling lakes are located at low altitudes. In contrast, the sampling lakes of groups III and IV are situated at high attitudes with low MAT and MAP. This fact is also supported by the PC1 and RDA1 axes, although the environmental implications of the PC2 and RDA2 axes are yet not clear. This suggests that the PC1 and RDA1 of phytoliths are comprehensive embodiments of MAT and MAP. Modern lacustrine phytolith assemblages can, thus, be used as indicators of climate in western Yunnan.
Some studies seem to support this conclusion of our study. Fredlund and Tieszen [81] showed that phytolith assemblages are likely to reflect the regional climate, and grasses are closely associated with temperature and moisture at a regional scale. In their study of phytoliths from modern soils in Eurasia, Lu et al. [44] found that MAP is the most important parameter controlling the spatial distribution of soil phytoliths. Liu et al. [78] found that increases in MAT resulted in higher production of Rondel and Saddle phytolith types, while production of Bulliform and long cell phytoliths was positively correlated with MAP. Phytolith production can vary with respect to climate. The distribution of phytoliths in our dataset in western Yunnan is significantly correlated with MAT and MAP. The relationship between phytolith assemblages and MAP has been documented previously [80,81,82], and the influence of MAT on the formation of phytoliths has also been well documented [83].
This study shows that phytoliths are abundant, widely dispersed, and well preserved in the lacustrine sediments of western Yunnan. And western Yunnan has complex terrain [13], a large latitude span [24], and diverse vegetation [27] and climate conditions [24]. Thus, this study’s findings have potential applicability to different regions. Therefore, modern lacustrine phytoliths can provide ubiquitous, climatically sensitive, and stratigraphically complete materials for paleovegetation and paleoclimate reconstructions. Since herbs are sensitive to climate and respond quickly to climate change, phytolith–climate transfer functions may offer reliable, high-quality climate reconstructions. Such data are essential for the development and assessment of climate models [84,85,86] and biodiversity conservation strategies [87].

6. Conclusions

Phytolith analysis of lacustrine sediments from 70 lakes in western Yunnan was conducted. PCA and RDA of phytolith data and climate data were also performed to reveal the spatial pattern of modern phytolith distribution, as well as the associations of phytoliths with vegetation and climate. Our study reached the following conclusions:
(1)
Modern lacustrine phytoliths in western Yunnan are dominated by herbaceous phytoliths, especially Poaceae types.
(2)
The 70 samples taken can be divided into 4 groups. Samples in group I and II are mainly distributed across vegetation zones A and B, and samples in groups III and IV across vegetation zones C and D, suggesting the association of modern lacustrine phytoliths and regional vegetation in the lake catchments.
(3)
The PCA and RDA results suggest that modern lacustrine phytolith distribution in western Yunnan was controlled by MAT and MAP. Phytolith–MAT/MAP transfer functions developed using WA-PLS (the best model among six models) and a training dataset consisting 50 samples in western Yunnan exhibited good performance, suggesting the associations of modern lacustrine phytoliths with regional climatic parameters in the lake catchments.
(4)
This study provides a basis for the vegetational interpretation and climate reconstruction of fossil phytolith records from lakes in western Yunnan.
(5)
Our results also offer a possibility for reliable high-quality climate reconstructions, which are necessary to develop and assess climate models and biodiversity conservation strategies.

Author Contributions

Conceptualization, C.S.; methodology, Y.X., C.S., M.W., Q.S. and H.S.; software, Y.X., M.W. and Q.S.; validation, Y.X., H.M. and L.H.; formal analysis, Y.X. and M.W.; investigation, Q.S., H.M. and L.H.; resources, Q.S., H.M., L.H. and H.S.; data curation, Y.X., H.M., Q.S., L.H. and H.S.; writing—original draft preparation, Y.X. and M.W.; writing—review and editing, C.S. and M.W.; visualization, Y.X.; supervision, C.S.; project administration, C.S.; funding acquisition, C.S., H.M., M.W. and H.S. All authors have read and agreed to the published version of the manuscript.

Funding

This research was funded by the National Natural Science Foundation of China: 42177437, 42167065, 41971115, 41372191, 41761044; the Special Project for Basic Research of Yunnan Province—Key Project: 202101AS070006; the Youth Talent Support Program of Xingdian Talent Plan, Yunnan Province: XDYC-QNRC-2022-0029; the Yunnan Project for the Introduction of Advanced Talents: 2013HA024; the Caiyun post-doctoral innovation project; the Yunnan Normal University Postdoctoral Research Project; the Yunnan Normal University Faculty of Geography Postdoctoral Fund: YNNU-FG-201; and the Yunnan Normal University Faculty of Geography Open Fund: YNNU-FG-202.

Data Availability Statement

The datasets used and generated in this study are available from the corresponding author upon reasonable request.

Acknowledgments

We thank three anonymous reviewers for their constructive comments and suggestions. We thank Kenneth Z. Shen for his helpful review and English improvement of the manuscript. We also thank all members of the 2017–2018 field trip groups for their help.

Conflicts of Interest

The authors declare no conflicts of interest.

References

  1. Smol, J.P.; Birks, H.J.B.; Last, W.M. Tracking Environmental Change Using Lake Sediments: Volume 3: Terrestrial, Algal, and Siliceous Indicators; Kluwer Academic Publishers: New York, NY, USA, 2001. [Google Scholar]
  2. Li, R.; Fan, J.; Vachula, R.S.; Tan, S.; Qing, X. Spatial distribution characteristics and environmental significance of phytoliths in surface sediments of Qingshitan Lake in Southwest China. J. Paleolimnol. 2019, 61, 201–215. [Google Scholar] [CrossRef]
  3. Piperno, D.R. Aboriginal agriculture and land usage in the Amazon Basin, Ecuador. J. Archaeol. Sci. 1990, 17, 665–677. [Google Scholar] [CrossRef]
  4. Piperno, D.R. Phytolith Analysis: An Archaeological and Geological Perspective; Academic Press: San Diego, CA, USA, 1988. [Google Scholar]
  5. Lu, H.; Wu, N.; Nie, G.; Wang, Y. Phytolith in loess and its bearing on paleovegetation. In Loess, Environment and Global Change; Liu, D.S., Ed.; Science Press: Beijing, China, 1991; pp. 112–123. [Google Scholar]
  6. Wen, C.; Lu, H.; Zuo, X.; Ge, Y. Advance of research on modern soil phytolith. Sci. China Earth Sci. 2018, 61, 1169–1182. [Google Scholar] [CrossRef]
  7. Wang, Y.; Lu, H. Phytolith Study and Its Application; China Ocean Press: Beijing, China, 1993. (In Chinese) [Google Scholar]
  8. Piperno, D.R. Phytoliths. In Tracking Environmental Change Using Lake Sediments: Volume 3: Terrestrial, Algal, and Siliceous Indicators; Smol, J.P., Birks, H.J.B., Last, W.M., Eds.; Kluwer Academic Publishers: New York, NY, USA, 2001; pp. 235–252. [Google Scholar]
  9. Rashida, I.; Mir, S.H.; Zurrob, D.; Dar, R.A.; Reshi, Z.A. Phytoliths as proxies of the past. Earth-Sci. Rev. 2019, 194, 234–250. [Google Scholar] [CrossRef]
  10. Qader, W.; Mir, S.H.; Meister, J.; Dar, R.A.; Madella, M.; Rashid, I. Sedimentological perspective on phytolith analysis in palaeoecological reconstruction. Earth-Sci. Rev. 2023, 244, 104549. [Google Scholar] [CrossRef]
  11. Jordi, L.P.; Zhang, F.; Ge, S. Plant biodiversity in China: Richly varied, endangered, and in need of conservation. Biodiv. Conserv. 2006, 15, 3983–4026. [Google Scholar]
  12. Mittermeier, R.A.; Turner, W.R.; Larsen, F.W.; Brooks, T.M.; Gascon, C. Global biodiversity conservation: The critical role of Hotspots. In Biodiversity Hotspots: Distribution and Protection of Conservation; Zachos, F.E., Habel, J.C., Eds.; Springer: Berlin/Heidelberg, Germany, 2011. [Google Scholar]
  13. Wu, S.; Pan, T.; Cao, J.; He, D.; Xiao, Z. Barrier-corridor effect of longitudinal range-gorge terrain on monsoons in Southwest China. Geogr. Res. 2012, 31, 1–13. (In Chinese) [Google Scholar]
  14. Chen, C.; Huang, R.; Yuan, D.; Zhang, J.; Cheng, H.; Ning, Y.; Yu, T.; Shen, C.; Edwards, R.L.; Long, X.-Y.; et al. Karst hydrological changes during the Late-Holocene in Southwestern China. Quat. Sci. Rev. 2021, 258, 106865. [Google Scholar] [CrossRef]
  15. Shen, J.; Jones, R.T.; Yang, X.; Dearing, J.A.; Wang, S. The Holocene vegetation history of Lake Erhai, Yunnan province southwestern China: The role of climate and human forcings. Holocene 2006, 16, 265–276. [Google Scholar] [CrossRef]
  16. Xiao, X.; Haberle, S.G.; Shen, J.; Yang, X.; Han, Y.; Zhang, E.; Wang, S. Latest Pleistocene and Holocene vegetation and climate history inferred from an alpine lacustrine record, northwestern Yunnan Province, southwestern China. Quat. Sci. Rev. 2014, 86, 35–48. [Google Scholar] [CrossRef]
  17. Shi, Q.; Shen, C.; Meng, H.; Huang, L.; Sun, Q. A 1640-Year Vegetation and Fire History of the Lake Haixihai Catchment in Northwestern Yunnan, Southwest China. Forests 2023, 14, 990. [Google Scholar] [CrossRef]
  18. Fearn, M.L. Phytoliths in sediment as indicators of grass pollen source. Rev. Palaeobot. Palynol. 1998, 103, 75–81. [Google Scholar] [CrossRef]
  19. Aleman, J.C.; Canal-Subitani, S.; Favier, C.; Bremond, L. Influence of the local environment on lacustrine sedimentary phytolith records. Palaeogeogr. Palaeocli. Palaeoecol. 2014, 414, 273–283. [Google Scholar] [CrossRef]
  20. Plumpton, H.J.; Mayle, F.M.; Whitney, B.S. Paleoecological potential of phytoliths from lake sediment records from the tropical lowlands of Bolivia. Rev. Palaeobot. Palynol. 2020, 275, 104113. [Google Scholar] [CrossRef]
  21. Pan, T.; Wu, S.; He, D.; Dai, E.; Liu, Y. Ecological effects of longitudinal range-gorge land surface pattern and its regional differentiation. Acta Geogr. Sin. 2012, 67, 13–26. (In Chinese) [Google Scholar]
  22. Luo, R.; Chen, X.; Xiong, L. Spatial distribution pattern and its causes of national forests and villages in Yunnan province. Chin. J. Agric. Resour. Reg. Plan. 2023, 44, 146–158. (In Chinese) [Google Scholar]
  23. Zhang, W.; Wan, Y.; Xiao, Z. The spatio-temporal variations of the longitudinal range-gorge region (LRGR) precipitation in recent 100 years in southeast China. J. Nat. Resour. 2006, 21, 802–809. (In Chinese) [Google Scholar]
  24. Cao, J.; He, D.; Yao, P. Research on the spatial distribution of rainfall and temperature in winter and summer over longitudinal range-gorge region (LRGR). Adv. Earth Sci. 2005, 20, 1176–1182. (In Chinese) [Google Scholar]
  25. Wang, Y. Yunnan Mountain Climate; Yunnan Scientific Publishing Press: Kunming, China, 2006. (In Chinese) [Google Scholar]
  26. Wu, Z.; Zhu, Y. Yunnan Vegetation; Science Press: Beijing, China, 1987. (In Chinese) [Google Scholar]
  27. Editorial Board of Vegetation Map of China, Chinese Academy of Sciences. Vegetation Atlas of China; Science Press: Beijing, China, 2001. (In Chinese) [Google Scholar]
  28. Yang, Y.G. About the principles and units of vegetational regionalism in Yunnan. Acta Bot. Yun. 1980, 2, 441–450. (In Chinese) [Google Scholar]
  29. Piperno, D.R. Phytoliths: A Comprehensive Guide for Archaeologist and Paleoecologists; AltaMira Press: New York, NY, USA, 2006. [Google Scholar]
  30. Ter Braak, C.J.F.; Prentice, I.C. A theory of gradient analysis. Adv. Ecol. Res. 1988, 18, 271–317. [Google Scholar]
  31. Birks, H.J.B. Quantitative palaeoenvironmental reconstructions. In Statistical Modelling of Quaternary Science Data. Technical Guide, Vol. 5.; Maddy, D., Brew, J.S., Eds.; Quaternary Research Association: Cambridge, UK, 1995; pp. 161–254. [Google Scholar]
  32. Kitao, A. Principal component analysis and related methods for investigating the dynamics of biological macromolecules. J 2022, 5, 298–317. [Google Scholar] [CrossRef]
  33. Hall, P.; Wilson, S.R. Two guidelines for bootstrap hypothesis testing. Biometrics 1991, 47, 757–762. [Google Scholar] [CrossRef]
  34. Juggins, S. User Guide: C2, Software for Ecological and Palaeoecological Data Analysis and Visualization User guide Version 1.3; Department of Geography University of Newcastle: Newcastle upon Tyne, UK, 2003; pp. 1–69. [Google Scholar]
  35. Ter Braak, C.J.F.; Juggins, S. Weighted averaging partial least squares regression (WA-PLS): An improved method for reconstructing environmental variables from species assemblages. Hydrobiologia 1993, 269, 485–502. [Google Scholar] [CrossRef]
  36. Birks, H.J.B. Numerical tools in palaeolimnology-progress, potentialities, and problems. J. Paleolimnol. 1998, 20, 307–332. [Google Scholar] [CrossRef]
  37. Ter Braak, C.J.F. Canoco-a FORTRAN Program for Canonical Community Ordination by (Partial) (Detrended) (Canonical) Correspondence Analysis, Principal Components Analysis and Redundancy Analysis (Version 2.1); TNO Inst: Wageningen, The Netherlands, 1988. [Google Scholar]
  38. Cameron, N.G.; Birks, H.J.B.; Jones, V.J.; Berge, F.; Catalan, J.; Flower, R.J.; Garcia, J.; Kawecka, B.; Koinig, K.A.; Marchetto, A.; et al. Surface-sediment and epilithic diatom pH calibration sets for remote European mountain lakes (AL: PE Project) and their comparison with the Surface Waters Acidification Programme (SWAP) calibration set. J. Paleolimnol. 1999, 22, 291–319. [Google Scholar] [CrossRef]
  39. Icpn Working Group: Madella, M.; Alexandre, A.; Ball, T. International Code for Phytolith Nomenclature 1.0. Ann. Bot. 2005, 96, 253–260. [Google Scholar] [CrossRef] [PubMed]
  40. Li, Q.; Lu, H.; Wang, W. Introduction and discussion on international code for phytolith nomenclature 1.0. Acta Palaeontol. Sin. 2009, 48, 131–138. (In Chinese) [Google Scholar]
  41. Mulholland, S.C.; Rapp, J.G. A morphological classification of grass silica-bodies. In Phytoliths Systematics: Emerging Issues Advances in Archaeological and Museum Science; Rapp, J.G., Mulholland, S.C., Eds.; Plenum Press: New York, NY, USA, 1992; pp. 65–89. [Google Scholar]
  42. Barboni, D.; Bremond, L. Phytoliths of East African grasses: An assessment of their environmental and taxonomic significance based on floristic data. Rev. Palaeobot. Palynol. 2009, 158, 29–41. [Google Scholar] [CrossRef]
  43. Wang, W.; Liu, J.; Zhou, X. Climate indexes of phytoliths from Homo erectus’ cave deposits in Nanjing. Chin. Sci. Bull. 2003, 48, 2005–2009. [Google Scholar] [CrossRef]
  44. Lu, H.; Wu, N.; Yang, X.; Jiang, H.; Liu, K.B.; Liu, T.S. Phytoliths as quantitative indicators for the reconstruction of past environmental conditions in China I: Phytolith-based transfer functions. Quat. Sci. Rev. 2006, 25, 945–959. [Google Scholar] [CrossRef]
  45. Lu, H.; Wu, N.; Liu, K.B.; Jiang, H.; Liu, T.S. Phytoliths as quantitative indicators for the reconstruction of past environmental conditions in China II: Palaeoenvironmental reconstruction in the Loess Plateau. Quat. Sci. Rev. 2007, 26, 759–772. [Google Scholar] [CrossRef]
  46. An, X.; Lu, H.; Chu, G. Surface soil phytoliths as vegetation and altitude indicators: A study from the southern Himalaya. Sci. Rep. 2015, 5, 15523. [Google Scholar] [CrossRef] [PubMed]
  47. Borcard, D.; Gillet, F.; Legendre, P. Numerical Ecology with R, (Use R!); Springer Science+Business Media: New York, NY, USA, 2011. [Google Scholar]
  48. ter Braak, C.J.F.; Verdonschot, P.F.M. Canonical correspondence analysis and related multivariate methods in aquatic ecology. Aquat. Sci. 1995, 57, 255–289. [Google Scholar] [CrossRef]
  49. Efremov, I.A. Taphonomy: A new branch of paleontology. Pan-Am. Geol. 1940, 74, 81–93. [Google Scholar]
  50. Piperno, D.R. Phytolith analysis and tropical paleo-ecology: Production and taxonomic significance of siliceous forms in new world plant domesticates and wild species. Rev. Palaeobot. Palynol. 1985, 45, 185–228. [Google Scholar] [CrossRef]
  51. Madella, M.; Lancelotti, C. Taphonomy and phytoliths: A user manual. Quat. Int. 2012, 275, 76–83. [Google Scholar] [CrossRef]
  52. Osterrieth, M.; Madella, M.; Zurro, D.; Alvarez, M.F. Taphonomical aspects of silica phytoliths in the loess sediments of the Argentinean Pampas. Quat. Int. 2009, 193, 70–79. [Google Scholar] [CrossRef]
  53. Latorre, F.; Honaine, M.F.; Osterrieth, M.L. First report of phytoliths in the air of Argentina. Aerobiologia 2012, 28, 61–69. [Google Scholar] [CrossRef]
  54. Crifò, C.; Strömberg, C.A.E. Small-scale spatial resolution of the soil phytolith record in a rainforest and a dry forest in Costa Rica: Applications to the deep-time fossil phytolith record. Palaeogeogr. Palaeocli. Palaeoecol. 2019, 537, 109107. [Google Scholar] [CrossRef]
  55. Aleman, J.C.; Blarquez, O.; Bentaleb, I.; Bonté, P.; Brossier, B.; Carcaillet, C.; Gond, V.; Gourlet-Fleury, S.; Kpolita, A.; Lefèvre, I.; et al. Tracking land-cover changes with sedimentary charcoal in the Afrotropics. Holocene 2013, 23, 1853–1862. [Google Scholar] [CrossRef]
  56. Albert, R.M.; Weiner, S.; Bar-Yosef, O.; Meignen, L. Phytoliths in the middle palaeolithic deposits of Kebara Cave, Mt Carmel, Israel: Study of the plant materials used for fuel and other purposes. J. Archaeol. Sci. 2000, 27, 931–947. [Google Scholar] [CrossRef]
  57. Tsartsidou, G.; Lev-Yadun, S.; Efstratiou, N.; Weiner, S. Ethnoarchaeological study of phytolith assemblages from an agro-pastoral village in Northern Greece (Sarakini): Development and application of a phytolith difference index. J. Archaeol. Sci. 2008, 35, 600–613. [Google Scholar] [CrossRef]
  58. Cabanes, D.; Weiner, S.; Shahack, G.R. Stability of phytoliths in the archaeological record: A dissolution study of modern and fossil phytoliths. J. Archaeol. Sci. 2011, 38, 2480–2490. [Google Scholar] [CrossRef]
  59. Koebernick, N.; Mikutta, R.; Kaiser, K.; Klotzbücher, A.; Nguyen, A.T.; Nguyen, M.N.; Klotzbücher, T. Controls on phytolith stability upon exposure in paddy soils. Geoderma 2024, 443, 116821. [Google Scholar] [CrossRef]
  60. Cabanes, D.; Shahack, G.R. Understanding fossil phytolith preservation: The role of partial dissolution in paleoecology and archaeology. PLoS ONE 2015, 10, e0125532. [Google Scholar] [CrossRef]
  61. Alexandre, A.; Meunier, J.D.; Colin, F.; Koud, J.M. Plant impact on the biogeochemical cycle of silicon and related weathering processes. Geochim. Cosmochim. Acta 1997, 61, 677–682. [Google Scholar] [CrossRef]
  62. Farmer, V.C.; Delbos, E.; Miller, J.D. The role of phytolith formation and dissolution in controlling concentrations of silica in soil solutions and streams. Geoderma 2005, 127, 71–79. [Google Scholar] [CrossRef]
  63. Borrelli, N.; Alvarez, M.F.; Osterrieth, M.; Marcovecchio, J. Silica content in soil solution and its relation with phytolith weathering and silica biogeochemical cycle in typical argiudolls of the Pampean Plain, Argentina-a preliminary study. J. Soil. Sediment. 2010, 10, 983–994. [Google Scholar] [CrossRef]
  64. Prentice, A.J.; Webb, E.A. The effect of progressive dissolution on the oxygen and silicon isotope composition of opal-A phytoliths: Implications for palaeoenvironmental reconstruction. Palaeogeogr. Palaeocli. Palaeoecol. 2016, 453, 42–51. [Google Scholar] [CrossRef]
  65. Fraysse, F.; Pokrovsky, O.S.; Schott, J.; Meunier, J.D. Surface chemistry and reactivity of plant phytoliths in aqueous solutions. Chem. Geol. 2009, 258, 197–206. [Google Scholar] [CrossRef]
  66. Alexandre, A.; Meunier, J.D.; Lézine, A.M.; Vincens, A.; Schwartz, D. Phytoliths: Indicators of grassland dynamics during the late Holocene in intertropical Africa. Palaeogeogr. Palaeocli. Palaeoecol. 1997, 136, 213–229. [Google Scholar] [CrossRef]
  67. Alam, A.K.M.M.; Xie, S.C.; Wallis, L.A. Reconstructing late Holocene palaeoenvironments in Bangladesh: Phytolith analysis of archaeological soils from Somapura Mahavihara site in the Paharpur area, Badalgacchi Upazila, Naogaon District, Banglades. J. Archaeol. Sci. 2009, 36, 504–512. [Google Scholar] [CrossRef]
  68. Nogué, S.; Whicher, K.; Baker, A.G.; Bhagwat, S.A.; Willis, K.J. Phytolith analysis reveals the intensity of past land use change in the Western Ghats biodiversity hotspot. Quat. Int. 2017, 437, 82–89. [Google Scholar] [CrossRef]
  69. Yin, L.; Li, M. A study on the geographic distribution and ecology of C4 plants in China·C4 plant distribution in China and their relation with regional climatic condition. Acta Ecol. Sin. 1997, 17, 350–363. (in Chinese). [Google Scholar]
  70. Twiss, P.C. Predicted world distribution of C3 and C4 grass phytoliths. In Phytolith Systematics: Emerging Issues; Rapp., G., Mulholland., S.C., Eds.; Springer: New York, NY, USA, 1992; pp. 113–128. [Google Scholar]
  71. Blinnikov, M.S.; Bagent, C.M.; Reyerson, P.E. Phytolith assemblages and opal concentrations from modern soils differentiate temperate grasslands of controlled composition on experimental plots at Cedar Creek, Minnesota. Quat. Int. 2013, 287, 101–113. [Google Scholar] [CrossRef]
  72. Silantyeva, M.; Solomonova, M.; Speranskaja, N.; Blinnikov, M.S. Phytoliths of temperate forest-steppe: A case study from the Altay, Russia. Rev. Palaeobot. Palynol. 2018, 250, 1–15. [Google Scholar] [CrossRef]
  73. Hilbert, L.; Neves, E.G.; Pugliese, F.; Whitney, B.S.; Shock, M.; Veasey, E.; Zimpe, C.A.; Iriarte, J. Evidence for mid-Holocene rice domestication in the Americas. Nat. Ecol. Evol. 2017, 1, 1693–1698. [Google Scholar] [CrossRef] [PubMed]
  74. Sun, Q. Spatio-Temporal Changes of Vegetation, Climate, and Forest Fire since the Last Glacial Maximum and Their Driving Mechanisms in Typical Lake Basins of Yunnan. Ph.D. Thesis, Yunnan Normal University, Kunming, China, 2023. [Google Scholar]
  75. Fan, B.; Lu, H.; Li, Y.; Shen, C. A novel approach for quantitatively distinguishing between anthropogenic and natural effects on paleovegetation. PNAS Nexus 2024, 3, 135. [Google Scholar] [CrossRef] [PubMed]
  76. Watling, J.; Iriarte, J.; Mayle, F.E.; Schaan, D.; Pessenda, L.C.R.; Loader, N.J.; Street-Perrott, F.A.; Dickau, R.E.; Damasceno, A.; Ranzih, A. Impact of pre-Columbian geoglyph builders on Amazonian forests. Proc. Natl. Acad. Sci. USA 2017, 114, 1868–1873. [Google Scholar] [CrossRef]
  77. Watling, J.; Shock, M.P.; Mongeló, G.Z.; Almeida, F.O.; Kater, T.; De Oliveira, P.E.; Neves, E.G. Direct archaeological evidence for Southwestern Amazonia as an early plant domestication and food production centre. PLoS ONE 2018, 13, e0199868. [Google Scholar] [CrossRef]
  78. Blinnikov, M.S. Phytoliths in plants and soils of the interior Pacific Northwest, USA. Rev. Palaeobot. Palynol. 2005, 135, 71–98. [Google Scholar] [CrossRef]
  79. Morris, L.R.; West, N.E.; Ryel, R.J. Testing soil phytolith analysis as a tool to understand vegetation change in the sagebrush steppe and pinyon-juniper woodlands of the Great Basin Desert, USA. Holocene 2010, 20, 697–709. [Google Scholar] [CrossRef]
  80. Liu, L.; Jie, D.; Liu, H.; Gao, G.; Li, D.; Li, N.; Guo, J.; Qiao, Z. Assessing the importance of environmental factors to phytoliths of Phragmites communis in North-Eastern China. Ecol. Indic. 2016, 69, 500–507. [Google Scholar] [CrossRef]
  81. Fredlund, G.G.; Tieszen, L.L. Modern phytolith assemblages from the North American Great Plains. J. Biogeogr. 1994, 21, 321–335. [Google Scholar] [CrossRef]
  82. Lu, H.; Wu, N.; Liu, D.S.; Han, J.; Qin, X.; Sun, X.; Wang, Y. Seasonal climatic variation recorded by phytolith assemblages from the Baoji loess sequence in central China over the last 150,000 a. Sci. China Earth Sci. Ser. D 1996, 39, 629–639. [Google Scholar]
  83. Pearsall, D.M. Paleoethnobotany: A Handbook of Procedures, 2nd ed.; Academic Press: San Diego, Chile, 2000. [Google Scholar]
  84. Bova, S.; Rosenthal, Y.; Liu, Z.; Yan, M. Seasonal origin of the thermal maxima at the Holocene and the last interglacial. Nature 2021, 589, 548–553. [Google Scholar] [CrossRef] [PubMed]
  85. Dong, Y.; Wu, N.; Li, F.; Zhang, D.; Zhang, Y.; Shen, C.; Lu, H. The Holocene temperature conundrum answered by mollusk records from East Asia. Nat. Commun. 2022, 13, 5153. [Google Scholar] [CrossRef] [PubMed]
  86. Zhang, W.; Wu, H.; Cheng, J.; Geng, J.; Li, Q.; Sun, Y.; Yu, Y.; Lu, H.; Guo, Z. Holocene seasonal temperature evolution and spatial variability over the Northern Hemisphere landmass. Nat. Commun. 2022, 13, 5334. [Google Scholar] [CrossRef]
  87. Dawson, T.P.; Jackson, S.T.; House, J.I.; Prentice, I.C.; Mace, G.M. Beyond predictions: Biodiversity conservation in a changing climate. Science 2011, 332, 53–58. [Google Scholar] [CrossRef]
Forests 15 01163 g001
Figure 1. Location, topography, and climate of the study region. Dots indicate the spatial distribution of sampling lakes. (a) A topographic map showing the spatial extent of Yunnan, defined by black curves and the location of the study region; the arrows display the SASM and EASM (East Asian summer monsoon). (b,c) The regional patterns of MAT and MAP, defined by isotherms and isohyets.
Figure 1. Location, topography, and climate of the study region. Dots indicate the spatial distribution of sampling lakes. (a) A topographic map showing the spatial extent of Yunnan, defined by black curves and the location of the study region; the arrows display the SASM and EASM (East Asian summer monsoon). (b,c) The regional patterns of MAT and MAP, defined by isotherms and isohyets.
Forests 15 01163 g001
Forests 15 01163 g002
Figure 2. Distribution of vegetation and vegetation zones in the study region. Different black shapes indicate the locations of sampling lakes in vegetation zones A–D in Table 1.
Figure 2. Distribution of vegetation and vegetation zones in the study region. Different black shapes indicate the locations of sampling lakes in vegetation zones A–D in Table 1.
Forests 15 01163 g002
Forests 15 01163 g005
Figure 5. Biplot of PCA on phytolith percentage data of 15 major phytolith types forming surface sediments of 70 lakes in western Yunnan.
Figure 5. Biplot of PCA on phytolith percentage data of 15 major phytolith types forming surface sediments of 70 lakes in western Yunnan.
Forests 15 01163 g005
Forests 15 01163 g006
Figure 6. Biplot of RDA on phytolith percentage data of 15 major phytolith types and 4 climatic parameters for 70 lakes in western Yunnan.
Figure 6. Biplot of RDA on phytolith percentage data of 15 major phytolith types and 4 climatic parameters for 70 lakes in western Yunnan.
Forests 15 01163 g006
Forests 15 01163 g007
Figure 7. Scatter plots of predicted values versus observed values, and residuals versus observed values for WA-PLS model (Component 1); (a,b) for MAT, and (c,d) for MAP.
Figure 7. Scatter plots of predicted values versus observed values, and residuals versus observed values for WA-PLS model (Component 1); (a,b) for MAT, and (c,d) for MAP.
Forests 15 01163 g007
Forests 15 01163 g008
Figure 8. The spatial distribution of samples from different phytolith groups in vegetation zones A–D.
Figure 8. The spatial distribution of samples from different phytolith groups in vegetation zones A–D.
Forests 15 01163 g008
Table 1. Distribution of vegetation zones and dominant species of different vegetation zones in study area [26].
Table 1. Distribution of vegetation zones and dominant species of different vegetation zones in study area [26].
Vegetation ZonesDominant GrassDominant WoodCultivated Plant
A: northern tropical seasonal rainforests and semi-evergreen monsoon forestsScleria levis; Neyraudia reynaudiana; Woodwardia japonica; Thysanolaena maxima; Dicranopteris linearis; Juncus effususAntiaris toxicaria; Pouteria grandifolia; Ananum allum; Ficus altissima; Chukrasia tabularis; Pometia pinnata; Terminalia myriocarpa; Semecarpus albescens; Phoebe nanmuCocos nucifera; Areca catechu; Ananas comosus; Musa nana; Oryza sativa; Zea mays; Camellia sinensis; Saccharum officinarum
B: southern subtropical monsoonal evergreen broadleaved forestsMyrsins africana; M. semiserrata; Vaccinium bracteatumCastanopsis; Pinus kesiya var. langbianensis; Lithocarpus echinotholusOryza sativa; Zea mays; Triticum aestivum
C: northern subtropical semi-humid evergreen broadleaved forestsNeyraudia reynaudiana; Heteropogon; Glochidion velutinumCyclobalanopsis; Castanopsis; Pinus yunnanensis; Picea asperata; Abies fabri; Tsuga chinensisOryza sativa; Zea mays; Triticum aestivum; Juglans regia; Camellia sinensis
D: cold–temperate coniferous forests and meadowsKobresiaPicea asperata; Abies fabri
Table 2. Summary statistics for the first four axes of RDA on a dataset of 4 climatic parameters and 15 major phytolith types from 70 lakes in western Yunnan.
Table 2. Summary statistics for the first four axes of RDA on a dataset of 4 climatic parameters and 15 major phytolith types from 70 lakes in western Yunnan.
Axis 1Axis 2Axis 3Axis 4
Eigenvalues0.2450.0150.00130.0005
Species–environment correlations0.8770.0550.0480.019
Inter-set correlation of significant environmental variables with axes
MAP (mm) (annual mean precipitation)−0.895−0.1090.1080.418
MAT (°C) (annual mean temperature)−0.951−0.284−0.123−0.008
Pjan−0.335−0.775−0.1010.526
Altitude0.729−0.1830.578−0.319
Table 3. Performance statistics of six models for the training dataset consisting of 70 samples and 15 phytolith types. RMSEP and the adjusted coefficients of determination R2_Jack are calculated by leave-one-out cross-validation.
Table 3. Performance statistics of six models for the training dataset consisting of 70 samples and 15 phytolith types. RMSEP and the adjusted coefficients of determination R2_Jack are calculated by leave-one-out cross-validation.
ModelMethodR2_JackAve_Bias_JackMax_Bias_JackRMSEP
MATWAWA_Inv0.526−0.0054.4681.585
WA_Cla0.541−0.0124.3192.007
WATOL_Inv0.515−0.0094.5901.603
WATOL_Cla0.530−0.0214.5172.030
WA-PLSWA-PLS Component 10.526−0.0144.4771.584
WA-PLS Component 20.447−0.0284.7751.731
WA-PLS Component 30.414−0.0234.8971.794
WA-PLS Component 40.389−0.0134.8321.844
WA-PLS Component 50.389−0.0224.9101.845
IKRPCAR_010.0740.0445.3382.232
PCAR_020.4740.0054.7251.672
PCAR_030.4640.0054.8091.689
PCAR_040.444−0.0065.0271.725
PCAR_050.4150.01594.9561.776
MATMAT0.444−0.1235.5091.737
WMAT0.446−0.1315.5171.733
MLEst_ML0.462−0.0024.7601.897
LWWALWWA_Inv0.5140.0534.6051.606
LWWA_Cla0.4760.0995.0181.855
MAPWAWA_Inv0.407−0.117276.803202.661
WA_Cla0.4300.004375.54291.65
WATOL_Inv0.408−0.376273.423202.55
WATOL_Cla0.431−0.581367.447290.411
WA-PLSWA-PLS Component 10.4070.540277.34202.707
WA-PLS Component 20.3600.411282.905212.334
WA-PLS Component 30.3200.925337.585220.895
WA-PLS Component 40.2921.556351.118227.272
WA-PLS Component 50.2771.231351.218230.979
IKRPCAR_010.08911.459436.553252.667
PCAR_020.01018,470.189,642.895,201.6
MATMAT0.2601.001304.869226.676
WMAT0.2580.840306.579227.111
MLEst_ML0.36158.311270.681261.716
LWWALWWA_Inv0.3220.822304.567217.209
LWWA_Cla0.258−4.901355.785299.725
Table 4. Performance statistics of the WA-PLS model for the training dataset consisting of 50 samples and 15 phytolith types.
Table 4. Performance statistics of the WA-PLS model for the training dataset consisting of 50 samples and 15 phytolith types.
ModelMethodR2_JackAve_Bias_JackMax_Bias_JackRMSEP
MATWA-PLSWA-PLS Component 10.6670.0162.6370.960
WA-PLS Component 20.6020.0692.6311.100
WA-PLS Component 30.6010.0682.7931.110
WA-PLS Component 40.6080.0602.7441.096
WA-PLS Component 50.6170.0472.6931.078
MAPWA-PLS Component 10.6390.626191.892140.437
WA-PLS Component 20.6235.754189.986145.411
WA-PLS Component 30.5937.792206.084152.567
WA-PLS Component 40.5709.028220.130158.157
WA-PLS Component 50.5618.734214.726160.287
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

Share and Cite

MDPI and ACS Style

Xu, Y.; Shen, C.; Wang, M.; Meng, H.; Sun, Q.; Huang, L.; Sun, H. Modern Lacustrine Phytoliths and their Relationships with Vegetation and Climate in Western Yunnan, SW China. Forests 2024, 15, 1163. https://doi.org/10.3390/f15071163

AMA Style

Xu Y, Shen C, Wang M, Meng H, Sun Q, Huang L, Sun H. Modern Lacustrine Phytoliths and their Relationships with Vegetation and Climate in Western Yunnan, SW China. Forests. 2024; 15(7):1163. https://doi.org/10.3390/f15071163

Chicago/Turabian Style

Xu, Yanyan, Caiming Shen, Min Wang, Hongwei Meng, Qifa Sun, Linpei Huang, and Huiling Sun. 2024. "Modern Lacustrine Phytoliths and their Relationships with Vegetation and Climate in Western Yunnan, SW China" Forests 15, no. 7: 1163. https://doi.org/10.3390/f15071163

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Metrics

Back to TopTop