Geographical Distribution and Relationship with Environmental Factors of Paphiopedilum Subgenus Brachypetalum Hallier (Orchidaceae) Taxa in Southwest China

The determination of the geographic distributions of orchid species and their relationships with environmental factors are considered fundamental to their conservation. Paphiopedilum subgenus Brachypetalum is one of the most primitive, ornamental, and threatened groups of Orchidaceae. However, little is known about the distribution of Brachypetalum orchids and how they are influenced by environmental factors. In this study, we developed a database on the geographical distribution of Brachypetalum orchids based on a large-scale field investigation in the Guangxi, Guizhou, and Yunnan provinces of southwest China (2019–2020). Using this database, we first adopted the nonparametric Mann–Whitney U test to analyze the differences in the geographical distributions and growth environments of Brachypetalum orchids. In addition, we also used the method of principal component analysis (PCA) to explore distribution patterns of Brachypetalum orchids in relation to environmental factors (topography, climate, anthropogenic disturbance, productivity, and soil) in southwest China. Our results indicated that Brachypetalum orchid species were mainly distributed in the karst limestone habitats of southwest China. In general, there were 194 existing localities with the occurrence of seven target orchids in the investigated area. Of the discovered species in our study, 176 locations (~90.7%) were distributed primarily in the karst habitat. Among them, the range of 780–1267 m was the most concentrated elevation of Brachypetalum orchids. In addition, the findings also suggested that the distribution of Brachypetalum orchids in southwest China was relatively scattered in geographical space. However, the density of the distribution of Brachypetalum orchids was high, between 104° and 108° E and between 25° and 26° N. The results of the Mann–Whitney U test revealed that there are obviously different geographical distributions and growth environments of Brachypetalum in southwest China. More specifically, we found some extremely significant differences (p < 0.001) in elevation, mean diurnal range, precipitation of coldest quarter, solar radiation, and exchangeable Ca2+ between the provinces of southwest China. The PCA analysis revealed that elevation, solar radiation, temperature (mean diurnal range, annual temperature range) and precipitation (precipitation seasonality, precipitation of the warmest quarter) were found to be the most significant factors in determining Brachypetalum orchids’ distribution. These findings have implications in assessing conservation effectiveness and determining niche breadth to better protect the populations of these Brachypetalum orchid species in the future.


Introduction
Orchidaceae is an exceptionally diverse family of the monocotyledons, including about 25,000 species worldwide, belonging to five subfamilies and 880 genera [1]. Significantly, the basis of these data, we attempted to (1) explore the patterns of Brachypetalum species in southwest China ( Figure 1); (2) study the differences in the growing environments in geographical distribution sites of Brachypetalum species in southwest China; (3) identify the critical environmental variables constraining the distribution of Brachypetalum species in southwest China, and (4) evaluate the in situ conservation status of Brachypetalum species in southwest China.

Study Species
Paphiopedilum is a genus of flagship orchids and comprises some of the most popular orchids due to their beauty and rarity [29]. They are mainly distributed in areas from tropical Asia to the Pacific islands, with some species extending to subtropical areas [13]. Among them, southwest China is one of the distribution centers for this genus. Furthermore, the subgenus Brachypetalum is considered to be the more primitive type of the genus Paphiopedilum. In addition, the threatening situations of the subgenus Brachypetalum species are highly prevalent, such as Paphiopedilum armeniacum [30,31] and Paphiopedilum micranthum [32]. The problems of habitat fragmentation, unreasonable mining, and population decline of the Brachypetalum subgenus group are very prominent. We investigated seven species of the subgenus Brachypetalum that occurred in the wild of southwest China in this study. These species are Paphiopedilum micranthum, P. emersonii, P. bellatulum, P. malipoense, P. concolor, P. wenshanense, and P. armeniacum. In addition, because of the great ecological variation of Paphiopedilum malipoense, three varieties (Paphiopedilum malipoense var. jackii, Paphiopedilum malipoense var. hiepii, and Paphiopedilum malipoense var. angustatum) were classified as the protospecies of Paphiopedilum malipoense. According to the latest published version of The National Key Protected Wild Plants List [33], all these species were national first-class protected species, except for Paphiopedilum micranthum, which was placed on the second-class protected species list. For more detailed information on these species, please refer to Table 1. The scientific names of these species are consistent with the modern international databases (https://www.catalogueoflife.org/, accessed on 6 November 2021). Table 1. A checklist of the subgenus Brachypetalum from southwest China is presented based on recent field investigations of orchids. The Red List level was based on the global assessment scope of IUCN (https://www.iucnredlist.org/, accessed on 6 November 2021). The protected level was based on The National Key Protected Wild Plants List [33]. The

Study Area
Southwest China is a hotspot for biodiversity conservation, characterized by heterogeneous regional climates, complex topography, and rich flora [34]. The study area covers Guangxi, Guizhou, and Yunnan provinces in southwest China and occupies approximately 0.8 million square kilometers ( Figure 1). The study area is entirely coincident with the administrative boundary. This area has a subtropical monsoon climate with the mean annual temperature ranging from −2.9 • C to 23.3 • C and mean annual precipitation ranging from 583 mm to 2608 mm per year. Because southwest China is located in a low-latitude area with huge elevation differences, and under the direct influence of the Indian Ocean Warm Current, the variety and quantity of plants are abundant in this area [35,36]. Furthermore, the mixed evergreen-deciduous broadleaf forest is a typical and representative vegetation type in southwest China, and it is also the main vegetation type we investigated. This type of forest has a complicated community structure, high habitat heterogeneity, rich biodiversity, and high productivity [37].

Distribution of Target Orchids
We first collected and compiled the historical occurrence data for target orchid species in southwest China using an online database (http://www.cfh.ac.cn/, accessed on 15 September 2021; http://www.sp2000.org.cn/, accessed on 16 September 2021; http://ppbc. iplant.cn/, accessed on 15 September 2021) and digital herbarium (https://www.cvh.ac.cn/, accessed on 5 November 2021), as well as various types of literature available (Flora of China, Provincial flora, research papers, and scientific investigation reports on Maolan, Mulun, and other relevant nature reserves). In addition, to conduct a more targeted survey and assessment of the target species, we also consulted with some relevant insiders. Finally, we collected a total of 69 historical distribution records of the subgenus Brachypetalum. However, these distribution records often lack detailed and specific location information, especially herbarium records. Based on the collected species distribution and habitat information, we used quadrat and line-transect methods to investigate the target species in the field. The field survey lasted for 139 days and was conducted between June 2019 and November 2020. We obtained longitude and latitude data for target orchids' occurrence sites (point data) using Google Earth.

Environmental and Climatic Data
To explore the influences of environmental factors on the distribution of orchids of the subgenus Brachypetalum in southwest China, we obtained topographic, climate, anthropogenic, productivity, and soil data from WorldClim (https://www.worldclim.org/, accessed on 7 September 2021), Resource and Environment Science and Data Center, Chinese Academy of Sciences (RESDC) (https://www.resdc.cn/, accessed on 7 September 2021) and the National Tibetan Plateau Data Center (TPDC) (http://data.tpdc.ac.cn/, accessed on 7 September 2021) databases. The variables obtained are shown in Table S1. The geographical coordinate system of all environmental variables was GCS_WGS_1984. In addition, to avoid redundancy and multicollinearity of variables, a Spearman correlation analysis was conducted in R software. Furthermore, variables with a high correlation coefficient (|r| > 0.8) were not taken into account for subsequent analyses [38,39]. The Spearman correlation coefficient between each variable presented is in Table S2. Finally, we selected 27 environmental variables before conducting the analysis ( Table 2).

Differences of Environmental Factors among Species Distribution Sites
In comparisons of two independent samples where the result is not normally distributed and the samples are small, a nonparametric test is applicable. One popular and approved nonparametric test for comparing results between two independent groups is the Mann-Whitney U test. In addition, the Mann-Whitney U test, also known as the Wilcoxon rank sum test or the Mann-Whitney-Wilcoxon test, is designed to compare if there is a difference in the dependent variable for two independent groups [47]. In this paper, we analyzed the differences in environmental factors using the Mann-Whitney U test. This was carried out in R and based on the package 'ggpubr', version 0.4.0 [48].

Relationship between Species Distribution and Environmental Factors
As a first step, correlation analyses were conducted between geographical distribution data (latitude and longitude) of seven orchid species and the values for the 27 environmental variables (correlation coefficients are presented in Table S2). Then, we took into account the dimension reduction of these environmental variables in each classification. Accordingly, we applied a principal component analysis (PCA), implemented using the package 'factoextra' version 1.0.7 in R [49,50]. We tested for correlations between environmental factors and species' geographical distributions using correlation analysis in R. In addition, PCA allows us to summarize and visualize the information within a data set containing observations described by multiple quantitative variables. In other words, PCA reduces the dimensionality of multi-variate data to two or three main components, which can be visualized graphically, with a minimal loss of information [50]. In this study, we used correlation analysis and principal component analysis to identify the main environmental factors influencing the distribution patterns of Brachypetalum species. All statistical analyses in the study were conducted with R software version 4.1.0 (available from: https://www.r-project.org/, accessed on 6 September 2021).

Distribution Pattern of Brachypetalum Species in Southwest China
Through field surveys conducted in 2019 and 2020, we confirmed the total number of locations with the presence of each target orchid species in southwest China (Table S3). Ultimately, a total of 194 existing geographical distribution locations of seven target orchids of the subgenus Brachypetalum were recorded in 50 counties (districts/cities) of southwest China (Figure 1). This resulted in a better understanding of the distribution and conservation status of each species in southwest China. As a whole, the number of extant distribution sites for Brachypetalum species in southwest China was highest in the Guizhou province, with 105 in total ( Figure 2). The number of extant distribution sites was small in Guangxi and Yunnan provinces, with 44 and 45 respectively (Figure 2). For different Brachypetalum species, Paphiopedilum micranthum has shown a similar distribution pattern as Paphiopedilum bellatulum. However, Paphiopedilum concolor, Paphiopedilum wenshanense, and Paphiopedilum armeniacum were mainly distributed in Guangxi and/or Yunnan. In addition, Paphiopedilum emersonii was mostly distributed in Guangxi and Guizhou provinces, whereas the Paphiopedilum malipoense was mostly distributed in Guangxi, Guizhou, and Yunnan provinces of southwest China (Figure 2). In general, the distribution pattern of Brachypetalum species in southwest China was relatively scattered and sparse. However, the density of Brachypetalum species distribution was high, between 104 • and 108 • E and between 25 • and 26 • N in southwest China ( Figure 3A). In addition, there was a very significant difference (p < 0.001) in the longitude of Brachypetalum species distributed in each province of southwest China ( Figure 3B). Furthermore, there was a very significant difference (p < 0.001) in the latitude of Brachypetalum species distribution between Guangxi and Guizhou provinces, as well as Guizhou and Yunnan provinces. However, no significant difference (p > 0.05) was observed in the latitude of Brachypetalum species distributed in Guangxi and Yunnan provinces ( Figure 3C).

Differences in Topographic Factors
The Mann-Whitney U test indicated that there was a very significant difference (p < 0.001) between the elevation of growing Brachypetalum species in each province ( Figure 4A). In terms of slope, there were some provinces with significant differences, except between Guangxi and Guizhou ( Figure 4B). However, concerning aspect, there was no significant difference between each province ( Figure 4C).

Differences in Productive Factors
The Wilcoxon rank sum test revealed extremely significant differences (p < 0.001) in the normalized difference vegetation index (NDVI) between Guangxi and Guizhou, as well as Guangxi and Yunnan ( Figure 5A). However, in terms of net primary production (NPP), there was only a significant difference (p < 0.01) between Guangxi and Guizhou ( Figure 5B).

Differences in Human Interference Factors
In consideration of multicollinearity, we only introduced population density (POP) as a factor. The results indicated that no significant difference (p > 0.05) was observed in population density among Guangxi and Yunnan. However, there were very significant differences (p < 0.001) in population density between the other two provinces and Guizhou, respectively ( Figure 6).

Differences in Climatic Factors
We found some significant differences (p < 0.001) in the mean diurnal range (Bio2) ( Figure 7B), precipitation of coldest quarter (Bio19) (Figure 7F), and solar radiation (Srad) ( Figure 7H) between the provinces of southwest China. Furthermore, the differences between provinces in terms of the climate factors mentioned above were higher than those in terms of the other climatic factors (Figure 7).

Differences in Edaphic Factors
Although there were some very significant differences (p < 0.001) between some provinces in terms of soil moisture (SMC), there was a minor difference (p < 0.05) between Guangxi and Guizhou ( Figure 8B). Similarly, we found that there were some very significant differences (p < 0.001) between some provinces in terms of the available K (AK), but there was a minor difference (p < 0.05) between Guizhou and Yunnan ( Figure 8I). The only extremely statistically significant difference (p < 0.001) that we found was in the area of exchangeable Ca 2+ (Ca 2+ ex ) ( Figure 8J).

Environmental Factors Related to Distribution Patterns of Brachypetalum Species
The greater the variable (distance from the origin of the axis), the more fully the variable is explained by the PC represented in that dimension (axis). Obviously, the first two dimensions explain roughly 57% to 71% of the variation in the data (Figure 9). In this research, principal component analysis showed that the distribution patterns of Brachypetalum species were highly related to the normalized difference vegetation index (NDVI) ( Figure 9B), elevation (Elev) (Figure 9C), average temperature (Tavg), solar radiation (Srad), water vapor pressure (Vapr), mean diurnal range (Bio2), annual temperature range (Bio7), precipitation seasonality (Bio15), precipitation of the warmest quarter (Bio18) ( Figure 9D) and exchangeable Ca 2+ (Ca 2+ ex ) ( Figure 9E).

Distribution Patterns of Brachypetalum Orchids in Southwest China
The orchid family is one of the richest in the realm of flowering plants and includes a lot of rare, endangered, or threatened species [4,51]. In this study, we discovered that the distribution of Brachypetalum orchid species is uneven in southwest China. More orchids occurred in the Guizhou province, an area of typical karstic geomorphology in China, whereas fewer orchids occurred in the Guangxi and Yunnan provinces. This may be due to uneven sampling efforts in each province. In addition, Yunnan province had the most abundant Brachypetalum orchid species, with six species in total, except for Paphiopedilum emersonii (Figures 1 and 2). Of course, Yunnan province was followed by Guangxi and Guizhou provinces, with four species of Brachypetalum orchids distributed in these two provinces, respectively (Figures 1 and 2). These findings showed that the distribution of Brachypetalum orchid species richness almost followed a general pattern of all wild orchids at the province-level scale in southwest China [52]. Furthermore, almost all orchid species were found in mountainous areas, which is consistent with the findings of Tang et al. [53], probably because the mountains provide heterogeneous habitats for various orchid species [2].
Some studies have focused on the ecological preferences of orchid species at regional scales. For example, Tsiftsis et al. [51] revealed the differentiation of the ecological or habitat preferences of some orchid species of east Macedonia. In that study, the vegetation types of the typical distribution sites of Brachypetalum orchid species were mixed evergreen/deciduous broadleaf forests. This also illustrated that certain habitat types or associated vegetation types were important determinants of orchid distribution [51]. Furthermore, the results of our study indicated that Brachypetalum orchid species were mainly distributed in the karst limestone habitats of southwest China, especially in southeast Yunnan province, Nu River in Yunnan province, southwest Guizhou province, southwest Guangxi province, and the junction between southern Guizhou province and northern Guangxi province (Figure 1). In general, there were 194 existing localities with an occurrence of seven target orchids in the investigated area. Of the discovered localities in our study, about 90.7% were distributed primarily in the karst habitat. They accounted for 176 locations from a total of 194 mapped locations. This may be related to the fact that most orchids belong to the group of calcific to basophilic species [54]. Therefore, it is sufficient to demonstrate that Brachypetalum orchids have strong habitat and ecological preferences for karst landform in southwest China. This is consistent with the findings of Wittlinger and Petrikovičová [54], who also agreed that the presence of orchids was concentrated primarily in a portion of the karst area.

Relationships between Environmental Factors and the Distribution of Brachypetalum Orchids in Southwest China
Orchids are usually sensitive to the environment, and their distribution is highly associated with environmental factors [55]. Based on some previous studies, some environmental factors (e.g., climate factors, soil factors, etc.) could influence orchid performance, thus impacting species distribution and population dynamics [3,51,56]. In this study, we explored the relationships between environmental factors and distribution patterns of Brachypetalum orchid species in southwest China. Moreover, we also found that many environmental factors played a significant role in the distribution of Brachypetalum orchid species in southwest China (Figure 9), which was consistent with the findings of previous studies [54].
According to previous research, human pressure has had an important impact on the distribution of some Cypripedium orchid species [8]. In this study, we also found that the population density (POP)-an index of human pressure or human activity-was a crucial factor in explaining the distribution of Brachypetalum orchid species in southwest China. The population density may affect the distribution of orchid species in the following two ways. First, high population density means an increase in the illegal harvesting of orchids, which can reduce the population of orchids. Furthermore, illegal harvesting has become a threat to the orchid species in southwest China, which may lead to a decrease in the number of recruits and adult individuals [57]. Second, high population density can increase human activity or human disturbance, such as grazing and deforestation, thus increasing fragmentation or habitat loss, which is the most serious threat to the survival of orchid populations. This may be due to human disturbances, which may lead to the breakdown of ecological connections between orchids and their mycorrhiza and pollinators [57].
Climate was an important factor affecting the distribution of Brachypetalum orchids in southwest China ( Figure 9D). In this study, we discovered that there were significant differences (p < 0.001) in the mean diurnal range (Bio2), precipitation of the coldest quarter (Bio19), and solar radiation (Srad) in the provinces of the study area. This indicated that the climatic niche of orchids distributed in various provinces is different, especially in terms of temperature, precipitation, and illuminance. This may be related to the difference between the growing altitudes of different orchid species, as altitude represents a complex variable associated with climate factors such as temperature and humidity [51]. Our study has scientific reference value for determining the ecological niche of Brachypetalum orchids in southwest China. Among the climatic factors, we revealed the relative significant roles of average temperature (Tavg), mean diurnal range (Bio2), annual temperature range (Bio7), precipitation seasonality (Bio15), precipitation of the warmest quarter (Bio18), solar radiation (Srad), and water vapor pressure (Vapr) in influencing the Brachypetalum orchid species in southwest China.
In addition to climate, productivity also contributed greatly to the distributions of Brachypetalum orchid species in southwest China. The productivity indexes of net primary production (NPP) and normalized difference vegetation index (NDVI) were important factors in explaining the distribution of Brachypetalum orchids in southwest China. This may be due to the fact that NDVI and NPP are likely to reflect the diversity and abundance of insects by providing food and habitats for a variety of insects [58,59], thus affecting the distribution of orchids, since most orchids are pollinated by particular insects [60][61][62].
In agreement with numerous previous studies on the distribution patterns of the various groups [63,64], we discovered that the elevation was very important in affecting the species distribution of Brachypetalum orchid species in southwest China. Other researchers have also reported that altitude was considered to be one of the most significant factors in determining orchid distribution [51]. Since the habitat requirement may vary by orchid species [65], areas with a large elevation range could provide various habitat conditions, and therefore contain more species. In addition, we found that many environmental factors changed with elevation in our research. For example, average temperature, water vapor pressure, precipitation of the coldest quarter, normalized difference vegetation index, and soil moisture decreased with elevation. Hence, the elevation may be one of the crucial factors structuring the distribution patterns of orchids in southwest China as it is linked to climate, productivity, and some soil properties [66].
According to the results of many previous studies [28,67], the majority of orchid species have a symbiotic relationship with mycorrhizal fungi. In turn, the distribution of mycorrhizal fungi depends to a large extent on soil conditions such as soil moisture, pH, organic content, and nutrients [3,28]. According to a previous study, soil factors (organic carbon, pH) had a significant impact on the distribution of some Cypripedium orchid species [8]. In this study, we found that soil condition is also one of the abiotic factors affecting the distribution of orchid species, which is consistent with the findings of Liu et al. [8]. On the one hand, this may be partly due to the fact that soil can influence the distribution of mycorrhizal fungi, thus affecting the distribution of orchids [3,28]. On the other hand, this is possibly because soil conditions can promote or inhibit the seed germination rate of orchid species [67,68]. Furthermore, exchangeable calcium ion (Ca 2+ ex ) has a significant influence on orchid species distribution ( Figure 9E), which coincides with the findings of Hrivnák et al. [69] and the essential role of calcium ions in pollen germination and pollen tube growth [70].

Conservation of Brachypetalum Orchids in Southwest China
The orchid family is a significant group with respect to biological conservation [11,64], because many are threatened with extinction. In situ conservation is very important in order to protect threatened species and populations in their natural habitat [25]. In this study, we discovered that there were 67 existing distribution locations of Brachypetalum orchids in the range of protected areas (Table S3). They accounted for 34.5% of the total 194 existing distribution locations, which indicated that there were still a large number of Brachypetalum orchids distributed outside the protected area without effective conservation. In addition, in the investigation, we found that unreasonable harvest and habitat destruction were the main threat factors of Brachypetalum orchids. In this regard, establishing protection plots can be an effective approach to the in situ conservation of Brachypetalum orchids in southwest China. In addition, actively developing ex situ conservation and wild reintroduction are also important to the conservation of orchids.
Moreover, identifying the niche of plant species is considered essential for their conservation [51]. In this study, we discovered that the elevation of Brachypetalum orchids ranged from 103.1 to 2024 m, with an average elevation of 974.4 m. Among them, the range of 780-1267 m was the most concentrated elevation of Brachypetalum orchids. Furthermore, all species occurred at this elevation range except Paphiopedilum armeniacum, which was distributed within a range of 1696-2024 m (Table S3). The results indicated that ecological preferences and niche breadth vary among species, which is consistent with Tsiftsis et al. [51], possibly because orchid germination niches are complex [71]. A better understanding of how niche breadth is associated with environmental gradients is crucial to species conservation, especially when considering the distribution of orchids [72]. These results revealed the niche breadth and the amplitude of geographical distribution of different Brachypetalum orchid species, which would be beneficial to set conservation priorities for the Brachypetalum orchid species in southwest China.

Conclusions
The geographical coordinates of Brachypetalum orchids obtained through this investigation strengthen our understanding of the distribution patterns and conservation status of each species in southwest China. Our findings showed that the distribution pattern of Brachypetalum species in southwest China was relatively scattered and sparse. In addition, Brachypetalum orchid species were mainly distributed in the karst limestone habitats of southwest China. In addition, more than 65% of Brachypetalum orchid sites were located outside protected areas and have not been effectively protected. Our preliminary but comprehensive non-parametric and correlative analysis implied that environmental factors play a significant role in determining the spatial distribution pattern of Brachypetalum orchids in southwest China. The PCA analysis also revealed that the distribution patterns of Brachypetalum orchids were most related to elevation, solar radiation, temperature (mean diurnal range, annual temperature range), and precipitation (precipitation seasonality, precipitation of the warmest quarter). Based on these results, we suggest that more indepth investigations of orchids' ecological preferences and niche breadth will be necessary to confirm its future conservation methods and programs, especially in the context of increasing anthropogenic disturbance and climate variability.
Supplementary Materials: The following are available online at https://www.mdpi.com/article/ 10.3390/d13120634/s1, Table S1: Environmental variables used to explain the geographical distribution of Brachypetalum orchids in southwest China.  Acknowledgments: The authors are very thankful to the editors and anonymous reviewers for their constructive comments and valuable suggestions on this manuscript.

Conflicts of Interest:
The authors declare no conflict of interest.