Multiperspective Analysis For Tree Ring Stable Carbon Isotope Variation Features

Stable isotopes in tree rings play an important and unique role on global change research in virtue of two great advantages——could provide palaeoclimate reconstructions with perfect annual resolution and statistically define confidence limits. In addition, trees are also widespread, so that it is possible to examine geographical variations in the climate of the past, which may be of more interest for predicting the consequences of future climate change than estimates of global or hemispherical conditions. Isotope ratios in tree rings have the added advantage that the physiological controls on their variation are reasonably well understood and relatively simple in comparison to the numerous factors controlling annual growth increment. The majority were restricted to the stable isotopes of carbon, the easiest and most rapid development stable isotope to measure. But until now there is no consensus yet with respect to the desirable composition, desirable sampling aspect and desirable material of tree ring carbon isotope ratio (δ13C) for better reflecting the climate and environment changes. It is necessary to perform comprehensive analysis between different compositions, different sampling aspects, different materials or different increment heights of tree ringδ13C of same species in the same study area. Unfortunately, the relative research has not been reported. This study aimed to investigate the distribution rule and variation pattern from above perspectives for the tree ringsδ13C of Pinus sylvestris var. mongolica Litv. which were collected from forest on the north slope of Yilehuli Mountian, Greater Khingan (appr.51°57’～52°00’N,124°22’～124°36’E), which is located in the exclusive cool temperature zone of China with coniferous forest vegetation. The main results are as follows:Firstly, the isotope signal of late wood in tree ring seems more sensitive to the change of local climate and environment than early wood. In some extent, the climate signal embedded in the holocellulose of tree ringδ13C is more significant than in theα-cellulose, which probably imply that part of the climate signal are lost during the extraction ofα-cellulose. The relationship betweenδ13C of whole wood with climate factors is statistically insignificant. This study regards that it is an acceptable way to use the composition of holocellulose in latewood for palaeoclimatical or palaeoenvironmental research.Secondly, the tree ringδ13C sequences show that the impact of regional climate and environment factors on the annual variation are identical. It is feasible that reconstruct the regional climate or environment change with tree ringδ13C sequences. There are noticeable differences in tree ring 813C with different aspects. There are not only most of positive correlations but part of negative correlations between the tree ringδ13C sequences. Specially, the sequences with significant correlation are usually exhibit symmetrical or obtuse-angled arrangement. The capability and extent of tree ring 813C withdrawing the climate factors vary widely among different aspects which accounts for the responses of different aspects to climate factors are incongruent. The annual variation for tree ringδ13C sequences could be represented by the annual variation of a certain amount of principal components.Thirdly, the general feature of tree ringδ13C sequences is as follows:The 813C values of transitional segment (TSG) are highest,δ13C values of early segment (ESG) come second andδ13C values of late segment (LSG) are lowest. The correspondences between the mean 813C value of the three materials and the relative indices obtained from vicinal meteorological stations reveal that the growing period could be separated by three stages: ESG (from the last third of April to the middle third of June, with greater soil moisture and faster increase in temperature), TSG (from the last third of June to the middle third of July, with lower soil moisture and temperature reaching highest level) and LSG (from the last third of July to the middle third of September, with greater soil moisture and fall of temperature) with low-high-low feature ofδ13C value. The seasonal trend ofδ13C is that fluctuating prominently from juvenile period to fast growing period and smoothing from maturation period to senile period. The amplitude of variation of LSG 813C is almost greater than ESG at the same period. Furthermore, the 813C of LSG is always significantly higher than ESG and they show synchronous change during juvenile period. The difference between ESG and LSG is taken off for maturation period and is negligible for senile period. The changes of annual 813C concentrate on the middle and later period of growing season and the role of TSG in the long-term 813C sequence could not be ignored.Furthermore, there are noticeable differences in tree ringδ13C with different height. Theδ13C increase significantly and reach the maximum under the shadow of tree crown, then increase and decrease once more along the stem toward the base. The change trend of average sequences of 813C are opposite with the average sequences of tree-ring width along the stem but parallel with the average sequences of the ratio of earlywood to latewood at the upper section of tree stem.The changing rate ofδ13C varied in different sections.All in all, the statistical results indicate that there are significant differences with varying degrees in each perspective of tree ring 813C sequences. The variation amplitudes are relative greater for materials and aspects and smaller for compositions and increment height in term of standard deviation. The difference among compositions is the most significant from the view of test statistics and the difference among aspects is more significant than the corresponding difference among years. The difference among different heights is close to the corresponding difference among years. The ANOVA analyses show that the differences in all studying perspectives of tree ring 813C sequences are almost of the same order of magnitude. The effect caused by different perspectives should be taken into consideration all around when focus on the paleoclimatical or paleoenvironmental research.