Variation in wood physical properties and effects of climate for different geographic sources of Chinese fir in subtropical area of China

Chinese fir is one of the most important commercial timber species in China, with many geographic sources. However, little is known of the variation in wood physical properties among them. To explore the differences in wood physical properties and their influencing factors, five geographic sources of Chinese fir were selected. The variance inflation factor, stepwise regression, and principle component analysis were used to reduce multicollinearity and dimensions of the 19 wood physical properties (including density, shrinkage, and mechanical properties). The results showed that the wood density differed significantly among five geographic sources. The tangential shrinkage rate and radial shrinkage rate reached maximum values in black-heart Chinese fir (HNYX-T) but accompanied by the lowest value for difference dry shrinkage. The wood density and mechanical properties of HNYX-T was exceeded to that of others geographic sources. Fast-growth Chinese fir (FJYK-P) had the lowest value for all mechanical properties. The precipitation and temperature had significant correlations with the wood physical properties of this five geographic sources. The temperature in summer was mainly positive correlated with physical properties, while precipitation was negatively correlated with them. HNYX-T had the highest comprehensive score of PCA, followed by JXCS-R, emerged as higher-quality geographic source, which is important for selecting and utilizing geographic sources in forest management.


Results
Variation in wood density. The values of Chinese fir's wood physical properties varied considerably among different geographic sources and Tukey-HSD testing showed that some of these differences were statistically significant (Fig. 1). The maximum value (HNYX-T) of wood all-dry density (WDD) was 62.70% higher than the minimum (FJYK-P). The WDD of each source was consistent with the classification and performance indexes of conifer trees in the timber strength grade for structural use, a standard in China's forestry industry 39 : FJYK-P was at level S10 (< 0.30 g/cm 3 ) and HNYX-T was at level S36 (< 0.50 g/cm 3 ).
The wood air-dried density (WAD) results matched the WDD, in that significant differences were found among all the geographic sources, with maximum value (HNYX-T) 60.85% higher than the minimum (FJYK-P) (Fig. 1a,b). The maximum value (HNYX-T) was 60.85% higher than the minimum value (FJYK-P). According to the classification regulations of wood quality in China 12 , the WAD of fast-growing Chinese fir was at the lowest level (≤ 0.35 g/cm 3 ). Likewise, according to the classification standard of physical properties indicators: FJYK-P was at level I (0.35 g/cm 3 ), while the other four geographic sources were at level II (0.35-0.55 g/cm 3 ). Through many experimental studies 19 , the Chinese Academy of Forestry concluded the WAD of Chinese fir in various regions ranged from 0.32 to 0.42 g/cm 3 . But here we found the HNYX-T (0.54 g/cm 3 ) and JXCS-R (0.49 g/cm 3 ) values exceeded 0.45 g/cm 3 .
Wood dry shrinkage was also an important indicator for evaluating its physical properties. The tangential shrinkage rate of all-dry (TSR.LD) of HNYX-P was the highest among the five geographic sources. The radial shrinkage rate of all-dry (RSR.LD) of HNYX-T was 67.70% higher than that of FJYK-P (2.10%). Among the five sources, the highest volume shrinkage rate of all-dry (VSR.LD) was obtained for HNYX-P, whereas the DDS.LD was the greatest in FJYK-P (2.85%), and was the lowest in JXCS-R (1.97%) ( Table 2). Variation in mechanical properties. As Table 3 shows, the modulus of rupture (MOR) of the five geographic sources was ranked as follows: JXCS-R > HNYX-T > HNYX-P > HNZJJ-P > FJYK-P. The flexural strength index was determined according to the China's classification standard of physical properties indexes. Both HNYX-T (110.70 MPa) and JXCS-R (95.60 MPa) were categorized as level III (88.10-118.00 Mpa); all other fir sources were designated level II (54.10-88.10 Mpa).
The modulus of elasticity (MOE) of HNYX-P was highest among the five geographic sources. Their ranking for tensile strength parallel to grain (TSG) was HNYX-T > JXCS-R > HNYX-P > HNZJJ-P > FJYK-P, for which the maximum was 47.60% higher than the minimum value. According to the grading standard of mechanical properties, HNYX-T, JXCS-R, and HNYX-P qualified for level III (10.4-13.2 GPa), while the other two sources were at level II (7.5-10.3 GPa).
The compression strength parallel to the grain (CSG) had this ranking: HNYX-T > JXCS-R, HNYX-P > HNZJJ-P > FJYK-P, for which the maximum 58.0% higher than the minimum value. According to the wood   Table 3). The compression strength perpendicular to the grain of total tensile (CPG.TT) among geographic sources was ranked as follows: HNYX-T > JXCS-R > HNYX-P > HNZJJ-P > FJYK-P (Table 4). Its maximum value (HNYX-T) was 29.3% higher than the minimum (FJYK-P). The ranking for compression strength perpendicular to the grain of total radial (CPG.TR) was slightly different: HNYX-T > JXCS-R > HNYX-P > HNZJJ-P > FJYK-P, for which the maximum was 42.1% higher than the minimum value. Compression strength perpendicular to the grain of part radial (CPG.PR) had the same rank order as CPG.TT, with a maximum value (HNYX-T) 35.0% higher than the minimum (FJYK-P). Finally, compression strength perpendicular to the grain of part tensile (CPG.PT) was ranked as HNYX-T > JXCS-R > HNZJJ-P > HNYX-P > FJYK-P for the five geographic sources of Chinese fir.

Factors influencing wood physical properties. Climate factors effect on wood physical properties. The
influence of precipitation on the three kinds of density was consistent. Pre in January, October, November, and December was positively related to wood density, while it was negatively correlated with density in others months, especially in May (r = − 0.39), June (r = − 0.59), and August (r = − 0.64). On a seasonal scale, Pre in summer was negatively correlated with density (r = − 0.77), but it was positively correlated with autumn (r = 0.22). MaxT was positively correlated with density during the whole year, except in May (r = − 0.34), and likewise with wood density but most strongly in summer (r = 0.75). MinT was positively correlated with density, especially in Jan (r > 0.7), though it was not significantly so in February and October (r < − 0.01). AveT was positively correlated with density except in January, February, and March, reaching statistical significance in June (r = 0.42), July (r = 0.55), and August (r = 0.64). AveT was positively correlated with density in all seasons except winter (r = − 0.12) (Fig. 2a).     MaxT showed no significant correlation with DDS.RD, RSR.LD, DDS.LD or VSR.LD (Fig. 2c). Pre had significant negative correlations with all of the mechanical properties in May, June, August, and summer, as evince by Fig. 2b, which also showed positive correlations in October. As we can seen, the effects of Pre on wood density and mechanical properties have the same tendency. Pre in all other months was not significantly correlated with mechanical properties (r < 0.3). AveT in January, February, March, and winter was negatively correlated with mechanical properties, but was positively correlated with mechanical properties in June, July, and summer, when the correlation coefficient reached its maximum, in August (r = 0.67). MinT was significantly correlated with mechanical properties, which was strongest in January (r > 0.75), while it was showed no significant correlation in Feb and Oct (r < 0.2). On a seasonal scale, MinT in winter was showed no significant correlation with mechanical properties (r < 0.3). As for MaxT, which was positively correlation with mechanical properties in August, October, and summer, while it was negatively correlation with mechanical properties in May, which was a interesting result we got from Fig. 2b. MinT showed a significant correlation with CSG, whose coefficient was higher 0.75 in summer. PCA was applied to the above 14 selected physical variables. These results showed that the physical properties of wood loaded strongly on the first axis of the PCA, explaining 51.8% of variation in the 14 tested properties, while the second axis explained 11.0% of it. MOE, MOR, TSR.RD, RSR.RD, and VSR.LD loaded on the positive axis of PC1 and PC2. Both DDS.LD and DDS.RD loaded on the negative axis of PC1 and PC2, while TSG, CSG, CPG.TT, CPG.TR, CPG.PT, CPG.PR, and WDD loaded on the positive axis of PC1 and the negative axis of PC2 (Fig. 3). For a comprehensive evaluation of Chinese fir's wood physical properties, we calculated the comprehensive scores of five geographic sources via the PCA. In this respect, significant differences were detected among the five geographic sources. Among them, the comprehensive score of HNYX-T was the highest whereas that of FJYK-P was the lowest (Fig. 4).

Discussion
Variation in physical properties. There were significant differences in wood density among the five geographic sources (P < 0.05). This was an expected result, and consistent with studies carried out elsewhere. For example, Luo et al. found that the wood density and mechanical properties of 32 Chinese fir clones differed significantly among them 36 , and wood density varies across globally among tropical tree species 41 . Importantly, the WBD values between 0.29 and 0.54 g/cm 3 in our study fell within the range (0.11-1.39 g/cm 3 ) reported for 2456 www.nature.com/scientificreports/ tropical forest tree species 42 . However, the average WBD (0.37 g/cm 3 ) in our study was lower than that (0.57 g/ cm 3 ) found for trees from two neotropical rain forests and seven subtropical tree species in China 43 . Compared with other species, Chinese fir is well known for its fast growth, which may explain its rather low density 7 . FJYK-P was the fastest growth geographic source in this study, perhaps because it has the lowest wood density, and shrinkage volumetric and radial values, and tangential and volumetric shrinkage coefficients. Wood geographic sources whose trees are of high-density harbor more dimensional variation, because they produce more wood per unit volume 44,45 . The mean value of volumetric shrinkage for the wood of Bracatinga (Mimosa scabrella Benthan) was 7.65% 46 , which is more than the values we obtained for Chinese fir woods. The eucalyptus tree (Eucalyptus benthamii Matenet Cambage) had radial shrinkage, tangential shrinkage, and an anisotropy coefficient of 5.91%, 13.87%, and 2.36 47 , respectively, which were almost the same as those found for the fir of five geographic sources in our study.
Through mechanical experiments carried out on 16 Chinese fir samples, Cai et al. found that in addition to the transverse compressive strength, the flexural strength, flexural elastic modulus, impact toughness, shear strength, and splitting strength were all higher than the radial surface, which also confirmed the accuracy of our conclusion 48 . The mean values of MOE, MOR, and CSG of fir from the five geographic sources in our study (Tables 3 and 4) were similar to those found in a previous study for L. sibirica that grows naturally in Mongolia, yet higher than those for L. kaempferi planted in Japan 49,50 . But the mechanical properties of HNYX-T are greater than other softwood species 51 .
The physical properties of wood determine its post-harvest applications. Among the five geographic sources, the comprehensive score was highest for black-heart Chinese fir, whose wood is generally used in the construction of ships, bridges, and as construction material 52 . Red-heart Chinese fir had many advantages, such as its round, straight trunk and fine grain, and being tough and corrosion resistant, with a large proportion of red heart wood, all of which generally make it popular in buildings and the furniture market 8 . Due to its growth rate and relatively low physical properties, fast-growing Chinese fir is generally marketable and widely in demand. We can see that the practical implications of Chinese fir's utilization were consistent with our wood property measurements.

Control factors for physical properties.
Trees and the climate factors are interdependent and are continually interacting. The growth and development of trees cannot be distinguished from the influence from climate conditions 53,54 . Our study showed that precipitation was significantly and negatively correlated with wood density in spring and summer (Fig. 2a). The possible reason for this finding is that abundant rainfall in summer causes trees to grow faster, resulting in a decreased density of formed wood, which is in line with a previous study 12 . In considering temperature only, a suitably high temperature can promote the growth of trees, but when that temperature limit is exceeded, it will inhibit the growth of trees.
The present study also revealed a significant and positive correlation of temperature with Chinese fir's mechanical properties (Fig. 2b). Similarly, according to Zhang et al. 's research, after eliminating the influence of evolutionary effects, the physical properties of wood was marked by a significant positive correlation with temperature. A positive correlation between temperature and MOE was also reported for Populus deltoides Marsh 50 .

Conclusion
By sampling trees and considering environmental factors, here we compared the wood physical properties of five different fir-growing geographic sources and the climate effects on physical properties. Our results showed that a certain degree of independence exists among the 19 wood physical properties. After calculated the comprehensive scores of five geographic sources via PCA, the HNYX-T emerged as the geographic source of highest-quality timber quality, which is important when evaluating geographic sources to select and utilize for growing Chinese fir. The influence of temperature was mainly positive with respect to physical properties of Chinese fir trees while precipitation was negatively correlated with them. But there were many additional factors likely affecting Chinese fir's physical properties, which could be promising for further research. Through the PCA of physical properties, the comprehensive score differed significantly among the five geographic sources, with HNYX-T having the highest score, followed by JXCS-R, HNYX-P, HNZJJ-P, and FJYK-P. The results provide a theoretical basis for future timber production and applications in forest management in subtropical China.

Materials and methods
Site description. Five geographic sources of Chinese fir samples in China were investigated: the fast growth Chinese fir samples were selected in Yangkou, Fujian Province (FJYK-P); the normal Chinese fir samples were selected in Zhangjiajie, Hunan Province (HNZJJ-P); the red-heart Chinese fir of samples were selected in Chenshan, Jiangxi province (JXCS-R); the black-heart Chinese fir (HNYX-T) and: normal Chinese fir (HNYX-P) samples were selected in the same county, which located in Yongshun, Hunan Province. All five sites are located in a subtropical region that has a moderate climate throughout the year and receives ample rainfall. During the year, high temperatures and much rainfall in these five areas are concentrated from June to September, while low temperatures and little rain begin in November and end in the following January (Table 5). (1) n = V 2 t 2 P 2 Table 5. The geographical and climate conditions of sampling sites. Sn is the number of test material samples, Pro is the name of province, SA is the average age of stand, SD is the stand density, Long is longitude, Lat is latitude, Alt is altitude, MAT is mean daily average temperature, MAP is the mean annual precipitation. where n is the number of samples (N), V is the coefficient of variation, t is the reliability index, P is the accuracy index.
The modulus of rupture and modulus of elasticity of the collected specimens, as well as their compression strength parallel to the grain, density, compression strength, and dry shrinkage test specimen, compression strength perpendicular to the grain (total tensile and total radial) and to the grain specimens (part tensile and part radial), and tensile strength parallel to the grain specimens were intercepted of blank in each one (Fig. 6).
Experimental methods and data source. The values of wood basic density (WBD, g/cm 3 ), wood alldry density (WDD, g/cm 3 ), wood air-dried density (WAD, g/cm 3  Temperature and precipitation data, which included average daily max temperature of each month (MaxT, °C), average daily min temperature of each month (MinT, °C), average daily mean temperature of each month (AveT, °C), and summed precipitation per month (Pre, mm) from 1969 to 2019 in four sample sites were obtained from Google Earth and the China Meteorological Data Sharing Service System (http://cdc.cma.gov.cn/ shishi/ climate.jsp). This data was classified on a seasonal basis, into spring (Mar, Apr, May), summer (Jun, Jul, Aug), autumn (Sep, Oct, Nov), and winter (Dec, Jan, Feb) for a given year. Data analysis. One-way analysis of variance (ANOVA) and the Tukey-HSD test were used to analyze differences of each wood physical property among the five geographic sources, at the confidence level of 0.95. Pearson correlations were used to analyze the relationship between climatic factors and the physical properties of Chinese fir. To resolve multicollinearity in the dataset, which consisted of 19 wood property variables, the variance inflation factor (VIF) was used 62 . Stepwise linear regression was used to filter the 19 wood property variables, and find out which variables contribute the most to the wood composite index (PCA composite score value) without multicollinearity. PCA was used as a method of dimension reduction transformation. All these statistical analyses were performed in R software v3.6.2 63 .

Data availability
Data could be shared directly through email to corresponding author email of Prof. Dr. Xiangwen Deng (dengxw@csuft.edu.cn).