Response of Korean pine’s functional traits to geography and climate

This study analyzed the characteristics of Korean pine (Pinus koraiensis) functional trait responses to geographic and climatic factors in the eastern region of Northeast China (41°–48°N) and the linear relationships among Korean pine functional traits, to explore this species’ adaptability and ecological regulation strategies under different environmental conditions. Korean pine samples were collected from eight sites located at different latitudes, and the following factors were determined for each site: geographic factors—latitude, longitude, and altitude; temperature factors—mean annual temperature (MAT), growth season mean temperature (GST), and mean temperature of the coldest month (MTCM); and moisture factors—annual precipitation (AP), growth season precipitation (GSP), and potential evapotranspiration (PET). The Korean pine functional traits examined were specific leaf area (SLA), leaf thickness (LT), leaf dry matter content (LDMC), specific root length (SRL), leaf nitrogen content (LNC), leaf phosphorus content (LPC), root nitrogen content (RNC), and root phosphorus content (RPC). The results showed that Korean pine functional traits were significantly correlated to latitude, altitude, GST, MTCM, AP, GSP, and PET. Among the Korean pine functional traits, SLA showed significant linear relationships with LT, LDMC, LNC, LPC, and RPC, and LT showed significant linear relationships with LDMC, SRL, LNC, LPC, RNC, and RPC; the linear relationships between LNC, LPC, RNC, and RPC were also significant. In conclusion, Korean pine functional trait responses to latitude resulted in its adaptation to geographic and climatic factors. The main limiting factors were precipitation and evapotranspiration, followed by altitude, latitude, GST, and MTCM. The impacts of longitude and MAT were not obvious. Changes in precipitation and temperature were most responsible for the close correlation among Korean pine functional traits, reflecting its adaption to habitat variation.


Introduction
By revealing plants' adaptation to changes in environmental conditions during their growth, functional traits provide a clear and convenient quantitative evaluation method [1], whose advantages are attractive for ecological studies at various scales. For example, some studies regulatory mechanism of Korean pine traits, we not only verified the applicability of plant functional trait theory but also provided an experimental basis for the study of the community structure and dynamics of Korean pine-dominated mixed coniferous broad-leaved forests.

Study sites
An overview of the geography of each sampling site is shown in Fig 1. Korean pine is mainly distributed in the eastern mountains of Northeast China (including Heilongjiang, Jilin, and Liaoning provinces), although it ranges between 34˚N and 52˚N and between 124˚E and 135˚E, along the eastern edge of the Eurasian continent. The northwestern side of this species distribution borders the Mongolian and the Siberian plateaus and its southeastern side is adjacent to the Japan, Bohai, and Yellow seas [42,43]. In this region, which is affected by the East Asian high-altitude monsoon climate and maritime air masses, the winter is cold and dry and the summer is hot and rainy. Because of geographic and climatic influences, in Northeast China, forest types change along a south-north gradient from warm-temperate deciduous broadleaf forest to temperate broadleaf and mixed forest, and to cold-temperate coniferous forest. From east to west, the area gradually changes from humid area to semi-humid and to semi-arid. The vegetation types are clearly distinct among forests and grasslands [44].
The eight sites where survey and sampling were conducted for the present study covered the main distribution areas of natural Korean pine (41˚16 0 N-48˚07 0 N; 124˚49 0 N-130˚02 0 E), being representative of its growth and development in Northeast China. These sites comprised nature reserves and forest areas with well-preserved Korean pine community structure and resources and with limited human interference, and an overview of the geography and climate of each sampling site is shown in Table 1. These regions preserved a relatively complete natural Korean pine community, suitable for carrying out studies on the effects of the geographic and climatic factors on Korean pine functional traits. Data Climatic data collection. Sampling sites were located in remote mountain forests in Northeast China lacking weather stations to collect accurate information. Therefore, regional climatic factors at different latitudes were calculated using a climate equation (see [40] for details), which was based on longitude, latitude, and altitude data to calculate temperature or precipitation:

Ethics statement
where T (or P) is the average monthly temperature (or precipitation), and a, b, c, and d are constants. This equation was developed based on the summarized relationship between geoclimatic information and forest vegetation types in China [40,45], and its applicability and reliability for estimating climate and precipitation in Northeast China has long been verified [31,40,46,47,48]. In the present study, longitude, latitude, and altitude data from the eight sampling sites were applied to the equation to obtain climate data. For analytical convenience, climatic factors were divided into three groups (Table 1): geographic factors, namely latitude, longitude, and altitude; temperature factors: mean annual temperature (MAT), growing season mean temperature (GST), and mean temperature of the coldest month (MTCM); and moisture factors: annual precipitation (AP), growing season precipitation (GSP), and potential evapotranspiration (PET). Trait measurement. Sampling was conducted from June to August 2015. A global positioning system (GPS) (UniStrong G120BD handheld GBS, Beijing, China) was used to collect longitude, latitude, and altitude information from each sampling site, and three replicates were taken from each sampling site. Ten to 20 plants were selected at each replication site, and their current-year leaves and underground fine roots were collected (for details of the functional traits measured and the methods used, see Table 2).
Data analysis. Data regarding Korean pine functional traits from the eight sampling sites were analyzed by SPSS 18.0 for Windows, using one-way analysis of variance (ANOVA) to determine whether differences among the mean values of traits at each sampling site were significant (P < 0.01). Values with significant differences were further examined by Duncan's consistency subset. Linear regression was performed using Origin pro 2016, with the least squares method, and all data for trait analysis were log 10    The Kjeldahl method was adopted for measuring these contents

Korean pine functional trait responses to latitude
Leaf phosphorus content LPC mgÁg -1 the molybdenum/antimony photometric method was adopted for measuring these contents Root nitrogen content RNC mgÁg -1 The Kjeldahl method was adopted for measuring these contents Root phosphorus content RPC mgÁg -1 the molybdenum/antimony photometric method was adopted for measuring these contents Some steps were followed before calculating the traits. (1) SLA, the pine leaves were divided into 5 equal parts (cylinders); the external surface (photoradiation surface) of each cylinder was defined as the leaf area. Leaves were scanned and leaf area was calculated using Image-Pro Plus 6.0 (Media Cybernetics, Bethesda, MD, USA). Leaves were then oven-dried at 65˚C for 72 h. (2) SRL, seedlings were cut at the transition between the stem and the root (where the tree bark texture changed). The roots were set aside for 24 h, dirt was washed off, and were then placed at room temperature for air-drying before scanning. Overlapping parts were split to facilitate scanning, and root length (sum of the taproot and lateral roots) was calculated using Image-Pro Plus 6.0. Roots were oven-dried at 65˚C for 72 h and weighed to obtain root dry weight.

Relationships of functional traits of Korean pine with geographic and climatic factors
The results of the linear correlation analysis between Korean pine and geographic and climatic factors are shown in Table 3. The number of Korean pine samples collected from the eight sampling sites varied between 139 and 240. The values of LT, LNC, LPC, RNC, RPC, and SRL were significantly correlated to latitude (P < 0.05), and LT, LPC, RPC, and SRL showed a negative linear relationship with latitude (R 2 ranged between 3% and 13%). Longitude showed a negative linear correlation with LDMC and SRL (P < 0.05), but did not show a linear relationship with other traits and R 2 was less than 3%. Altitude showed significant correlations with all traits (P < 0.05), and presented a negative linear relationship with LDMC (R 2 ranging between 3% and 21%). Although MAT showed a negative linear relationship with RNC, it had a positive linear relationship with RPC (P < 0.05) and an R 2 around 4%. Significant linear relationships were found between GST and SLA, LT, LDMC, LNC, LPC, and SRL (P < 0.05), with R 2 ranging between 2% and 10%. Although MTCM was positively correlated with LT, LNC, LPC, RNC, and RPC (P < 0.05), it was negatively correlated with LNC and RNC, with R 2 ranging between 2% and 11%. Except for SLA, all traits presented positive linear correlations with AP and GSP (P < 0.05), with R 2 ranging from 2-12% and 2-11%, respectively. Except for RNC, significant correlations were found between PET and all other traits (P < 0.05), and PET showed negative linear relationships with LT, LPC, RPC, and SRL, with R 2 ranging between 2% and 13%. The above analyses indicated that latitude, altitude, GST, MTCM, AP, GSP, and PET had obvious effects on Korean pinetraits.

Correlations in functional traits of Korean pine
In addition to the correlations with geographic and climatic factors, there were also correlations among Korean pine functional traits (Table 4). Results evidenced significant correlations among Korean pine traits (P < 0.001 or P < 0.05), and the mean R 2 value (0.116) indicated a low linear explanation ratio. The highest R 2 (0.216) was found between RNC and RPC, whereas the R 2 between SRL and RPC was only 0.038. Slope comparisons revealed that SLA and LT had a negative linear relationship with LDMC and LNC, LT showed a negative linear relationship with LNC and RNC, LDMC showed a negative linear relationship with LPC, and negative linear correlations were also found between LNC and LPC, RNC and RPC, LNC and RPC, and LPC and RNC. The remaining Korean pine traits showed positive linear relationships among them.

Effects of geographic and climatic factor variations on Korean pine functional traits
Our results showed that not all Korean pine functional trait responses to geographic and climatic factors were significant. Among the geographic factors, the effects of latitude and Korean pine functional responses to geography and climate altitude on Korean pine functional traits were obvious; however, longitude only correlated with LDMC and SRL, and its effects on other traits were not obvious. The reason might be that changes in temperature and precipitation were more obvious with increasing latitude and altitude than with changing longitude. Some studies have shown that changes in climatic conditions caused by changes in latitude are the main cause for changes in individual and community traits [26], which is in congruence with our results. A recent study showed that herbivory decreased with latitude and increased with temperature in the Northern Hemisphere [49], which is similar to the results obtained in our study. Some other studies suggest that both latitude and longitude cause changes in climatic conditions [27], which is not supported by our results. This might be because of the narrow longitude span of Korean pine in China, which does not allow the capturing of obvious variations in climatic factors. On the other hand, changes in the characteristics of Korean pine along its vertical distribution were obvious. It has been suggested that changes in temperature and precipitation resulting from increasing altitude were the main reason for Korean pine functional trait variations [31,38], which is consistent with the findings in the present study. Concurrently, previous studies also indicated that the main reasons for habitat heterogeneity were elevation and terrain, whereas the contribution from increasing latitude were not obvious [27,50]. The present study found that Korean pine functional traits were linearly correlated with both latitude and altitude, not ruling out the possibility of interaction between these two factors. However, further studies are required to determine which factor had a major impact. Moisture also affects the responses of Korean pine functional traits. Northeast China is affected by maritime air masses in the summer, and thus, has abundant rainfall. In the present study, GSP significantly correlated with all Korean pine traits except SLA, suggesting that this pine species is especially affected by moisture during its growth period. The limiting effects of precipitation and temperature on plant growth are key issues discussed in numerous studies. It has been suggested that in mid-latitude regions, the main limiting factor for plant growth and community distribution is precipitation; however, in high-latitude regions, the limiting effects of temperature are obvious [23]. The results of the present study are consistent with the above findings from previous studies.
Among temperature factors, GST and MTCM had clear effects on Korean pine, whereas the effect of MAT was not obvious. Because Northeast China is dually affected by high-altitude monsoons and maritime air masses and is subject to wide temperature variations throughout the year, changes in MAT might not explain the habitat conditions of Korean pine. It has been suggested that the lowest temperature is limiting for an organism survival [46], and GST in biomass accumulation is even more significant [5]. In fact, GST and MTCM reflect changes between the highest (summer) and the lowest (winter) temperatures in a year, and the results obtained in the present study also evidenced significant relationships between Korean pine functional traits and these two factors. A balanced condition between water dissipation and heat in a region is reflected in PET, which is closely related to MAT [40,51]. In the present study, PET showed significant correlations with all Korean pine traits except RNC, indicating that the degree of water evapotranspiration and heat level not only affected ecological regulation abilities such as absorption, transport, and nutrient and energy storage but also clearly affected the morphology of tissues and organs as well as their energy consumption. Both PET and MAT have been suggested to positively correlate with the leaf life span, which mainly explained the regional differences in functional traits [52].

Relationships among Korean pine functional traits
The correlation between Korean pine functional traits reflected its adopted ecological regulation strategy in response to changes in environmental conditions. Although SLA and LT showed a significant positive linear correlation, both showed a significant negative linear correlation with LDMC: as SLA decreased with increasing latitude, LT decreased and LDMC increased. These results indicated that Korean pine decreased leaf area and thickness while increasing leaf dry matter content to resist the negative impacts caused by increasing latitude and decreasing temperature. Some studies suggested a close relationship between SLA, LT, and LDMC [53,54]. The normal metabolic capacity is maintained by decreasing water loss during respiration and increasing leaf tissue density [14]. Although SLA significantly correlated with altitude, GST, and PET, no obvious correlation was found between SLA and longitude, latitude, and other temperature and precipitation factors, suggesting that the physiological maintenance mechanism adopted by Korean pine mainly responded to latitude changes. However, these changes aimed to cope with the changes in growth season temperature and potential evapotranspiration caused by regional altitude differences. Studies have used SLA to determine plant photosynthetic capacity and the potential growth rate [9,16], and temperature had an obvious limiting effect on plant photosynthesis and respiration [22]. In the present study, SLA responses to GST reflected the environmental adaptability of Korean pine in maintaining normal photosynthesis and respiration capacities during growth. Some studies have considered LDMC as the leaf construction cost and the ability of leaves to store energy and nutrients [52] and LT as the light acquisition ability of plants and the basis for determining the CO 2 diffusion pathway and resource utilization strategies [54]. Other studies used changes in LDMC and SLA as the basis for determining the ability of plants to absorb soil elements [55]. In the present study, Korean pine grew in the mid-latitude region, and SLA and LT decreased with decreasing temperature; therefore, photosynthetic and respiratory capacities decreased. It has been suggested that under such conditions, leaf life span increases, which is an adaption strategy of evergreen and coniferous species in response to adverse environments [1]. In addition, another study indicated that the interspecific level herbivory had a hump-shaped relationship with leaf lifespan and a positive relationship with leaf size, and nutritional traits have little relationship to herbivory [56]. This indicates that the relationship between functional traits is interrelated between species. The trends of changes in Korean pine leaf nitrogen and phosphorus elements were similar to those of roots, suggesting that their responses to latitude variation were similar, and there was a significant correlation between them. Nitrogen is a necessary material for the synthesis of photosynthetic elements, and both nitrogen and phosphorus provide energy and material for enzymatic photophosphorylation reactions [57]. In the present study, LNC showed a significant positive linear correlation with RNC, which explained root absorption, nitrogen transport, and leaf photosynthesis to provide raw materials; on the other hand, root growth and development requires material and energy, and thus roots have high nitrogen content. However, LNC showed a negative linear relationship with LPC and RPC, indicating that Korean pine reduced leaf and root energy consumption during respiration by decreasing phosphorus content with increasing latitude. In addition, LNC showed a negative linear relationship with SLA and LT, suggesting a reduced leaf area, which is considered a trait for species with a long leaf life span and low photosynthetic capacity [52]. It has been suggested that leaf nitrogen and phosphorus contents indicate soil nutrient limitation. The results of the present study suggest that phosphorus limitation is more pronounced in this area, and an increase in latitude did not result in decreased leaf and root nitrogen contents. This might be related the abundant rainfall and low mean temperature of Northeast China, which leads to nitrogen limitation and to a slow-growth strategy. Nitrogen, photosynthetic and respiratory capacities, and leaf morphology are related, and therefore, studies have linked plant's nitrogen-use-strategy with herbivore grazing to explain material circulation and energy loss in the food web [9].
Analyzing the relationship between functional traits and geographical and climatic factors of Korean pine also yielded reference values for the regeneration and protection of Korean pine species in northeast China. Unlike previous studies that have focused on individual growth and biomass accumulation, analyzing the functional traits in the present study allows emphasizing the limiting effect of external environmental conditions, such as different geographical factors, temperature, and precipitation, on the growth of Korean pine. Our results could help select the appropriate precipitation and altitude areas to perform manual regeneration, planting, and management of Korean pine, which can enhance the efficiency of the protection of Korean pine forest.

Conclusions
This study analyzed Korean pine functional trait responses to latitudinal variation in Northeast China and concluded that differences in Korean pine functional trait responses to latitude were mainly because of the effects of climatic factors resulting from geographic variations. Korean pine functional trait responses to AP, GSP, and PET were the most obvious, followed by responses to altitude and latitude, and finally to GST and MTCM. Differences in Korean pine functional traits at different latitudes were noteworthy and correlations between traits were significant, which reflect the ecological adjustment measures by Korean pine to adapt to changes in environmental conditions. Geographic and climatic factors clearly contributed to regional differences in Korean pine functional traits. Precipitation, temperature, evapotranspiration, and element contents were the main factors limiting Korean pine functional traits.
We studied the functional traits of Korean pine species, which, as mentioned above, is determined by the status of Korean pine in the forest community in Northeast China. Although the experimental results have certain value for the protection of Korean pine, the limitations are still evident in the validation of a functional trait theory. For example, the functional traits of a single species are not exactly the same as those of a community or ecosystem, and this problem will be analyzed in future studies.