Climatic signals in tree-ring width and stable isotopes composition of Pinus sylvestris L. Growing in the industrialized area nearby Kędzierzyn-Koźle

Kędzierzyn-Koźle (50°20 ′N; 18°13′E) is located near two factory complexes (Fig. 3): Blachownia Holding S.A. (previously named: Blachownia Chemical Factory), where chemical production began in 1941, and nitrogen factory Grupa Azoty ZAK (previously named: Zakłady Azotowe Kędzierzyn-Koźle, ZAK) where production started in 1954. In both factories, many different units and production facilities were constructed and added over a period of the last 60 years. Since the1980s, numerous projects dedicated to environmental protection were implemented in both factories. After many years of trying to minimize the negative impact of production processes on the environment, these factories were removed from the list of the most environmentally noxious companies. In 2004, the Waste Management Plan for the City of Kędzierzyn-Koźle set strategic objectives for 2015, inter alia: the introduction of the principles of “cleaner production”. The companies operating in Kędzierzyn-Koźle are of strategic importance to the region and the country.

Fig. 2

Fig. 3

The period from 1951 to 2012 AD was characterised in the regional climate records by an annual average temperature of about 9°C (data range from 6.7 to 10.6°C), and a mean annual sum of precipitation of around 610 mm (data range from 359 to 868 mm/year); mean annual number of sunshine hours is approx. 1530 (data range from 1108 to 1978), relative humidity around 80% (data range from 75 to 86%). The lowest precipitation was observed between mid-1980s and mid-1990s (Fig. 2). The vegetative period begins in April and lasts until September. The meteorological data were obtained thanks to the Polish Institute of Meteorology and Water Management (IMGW-PIB). Temperature, humidity and precipitation data comes from the meteorological station in Opole, whereas sunshine data come from the meteorological station in Katowice.

Scots pine (Pinus sylvestris L.) is a dominant species in the investigated area. For dendrochronological analysis, twenty trees were selected from each population (tree stands). The examined pines were healthy and grew in similar habitat conditions at different distances and directions from the industrial factories (Fig. 3).

All sites were classified as fresh mixed broadleaved forests. In order to avoid different dendroecological reaction of juvenile wood, an attempt was made to select pine stands aged between 80 (site A) and 100 years (sites C, B, N, S and T) which is the felling age of Scots pine. The pine trees were dominant or co-dominant individuals.

One increment core from each tree was taken at a height of 1.3 m above ground. Tree-ring widths were measured to the nearest 0.01 mm. The tree-rings were dated and rechecked using the COFECHA program (Holmes, 1983). Each tree-ring width series was standardized to remove non-climatic trends due to age and growth effects. Therefore, in each year the annual sensitivity index (s_i) was calculated according to the formula: Si=2⋅<!-- ? -->(xi−<!-- - -->xi−<!-- - -->1)⋅<!-- ? -->(xi+xi−<!-- - -->1)−<!-- - -->1$$\begin{array}{} \text{S}_{\text i}=2 \cdot (\text{x}_{\text i}-\text{x}_{{\text i}-1})\cdot (\text{x}_{\text i}+\text{x}_{{\text i}-1})^{-1} \end{array} $$(2.1)

where x_i is the tree-ring width in year i (Douglas, 1920).

The mean change of the tree-ring width for the 1951–2012 period was evaluated by the mean sensitivity (MS). MS indicates interannual changes of the trees’ sensitivity to climate factors (Fritts, 1976). The site sensitivity chronology was constructed on the basis of series sensitivity. The site sensitivity chronology exposed the short-term variance due to variation of climatic factors. The similarity of short-term incremental reactions of trees in each pine populations was evaluated by calculating the mean between-tree correlation (r_bt). The calculated index EPS enables an assessment of the representativeness of the constructed site chronologies (Wigley et al., 1984).

The principal component analysis (PCA) was applied to identify the short-time factors affecting tree-ring widths. The identification of PC1 and PC2 was based on an analysis of the component scores. The variables (n=52) were site sensitivity chronologies. The cluster analysis (CA) based on Ward’s method and 1-r Pearson’s distance has been used to analyse similarity of the response of each pine population to climate elements in the 1951–2012 period. The variables (n=52) were correlation coefficients between the site sensitivity chronologies and monthly temperature, precipitation, relative humidity and sunshine duration from the prior September to the current September. Response function analysis (Fritts, 1976; Holmes and Lough, 1999) was used to identify the climatic factors that determined the PC1 and PC2. These results were verified by analyzing the correlation between site sensitivity chronologies and the climate parameters.

The analysed samples covered the period 1975–2012 AD. The α-cellulose samples were extracted from increment cores of ten representative trees growing in the C sampling site. The pines growing in this sampling site were characterized by the highest level of tree ring width reduction in the period of time between 1958–1991.

The absolutely dated annual tree-rings were manually separated as thin slivers and pooled and homogenized. The α-cellulose samples were extracted by applying the procedures based on the Green’s method (1963) used in The Mass Spectrometry Laboratory of the Silesian University of Technology (Pazdur et al., 2007, 2013; Sensuła et al., 2011a, 2011b; Sensuła and Pazdur, 2013a, 2013b). This method includes the following steps: the removal of lignin, the processing of the holocellulose to α-cellulose, bleaching, neutralizing and drying (Green, 1963).

The initial stage of the process involved cutting of annual rings from the cores in the sample producing fine shavings <0.5 mm thick, cut from wood with a knife. Afterwards, the samples were treated with a toluene-ethyl alcohol mixture (in proportion 1:1, at 80°C, over 4 hours), ethyl alcohol (at 80°C, over 4 hours), distilled water (at 80°C, over 1 hour). The reaction was performed in a Soxhlet apparatus. Afterwards, the samples were dried overnight. The dried and weighed samples were placed into the glass test-tubes. The next step was bleaching with NaClO₂ and CH₃COOH solution (175 ml of distilled water, 2.5 g of NaClO₂ and 1.7 ml of CH₃COOH was added per 1g of each sample) at 70°C, over 1 hour in order to remove lignin and to extract cellulose. This process was repeated 5 times. In the next step the solution was removed by decanting. The samples were rinsed with boiling distilled water and then with cold distilled water up to neutral pH. Afterwards, the samples were treated with 50 ml of 10% NaOH solution (at 70°C, over 45 minutes). Afterwards the solution was removed and samples were rinsed with cold distilled water. Then the samples were treated with 17% NaOH solution (at room temperature for 45 minutes). The solution was then removed from the tubes and samples were rinsed with distilled water. At the end of the extraction process the samples were treated with 10 ml of 1% HCl solution and rinsed up to neutral pH. The obtained α-cellulose was dried on a hot plate at 60°C overnight. To ensure homogeneity the chemical pre-treatment was carried in an ultrasonic bath.

In order to determine the δ¹³C values, the samples (0.060 mg) were loaded into tin capsules and combusted at a temperature of 1100°C, CO₂ was separated in a gas chromatography column of the elemental analyser (EuroVector). In order to determine the δ¹⁸O values, the samples (0.095 mg) were loaded into silver capsules. To displace moisture-containing air in the cellulose samples, the samples were heated over 24 hours in a vacuum line (60°C) and after that they were put into a special air-filled dry box before stable isotope ratio determination (Sensuła et al., 2011b). The samples were converted to CO by pyrolysis at a temperature of 1350°C and separated in a gas chromatography column in the elemental analyser (EuroVector).

The stable oxygen and carbon isotope compositions of the samples were determined using an Isoprime continuous flow isotope ratio mass spectrometer (GV Instruments, Manchester, UK) at the Mass Spectrometry Laboratory of the Silesian University of Technology.

The relative deviation of the isotopic composition is expressed, in parts per thousand (‰), as δ=(R_sample /R_standard −1)⋅ 1000, Where R_sample and R_standard are the ratios of the heavy to the light isotope concentration in the sample and in the standard, respectively. The δ¹³C results are reported in values relative to VPDB (Vienna Pee Dee Belemnite), whereas the δ¹⁸O results are reported in values relative to VSMOW (Vienna Standard Mean Ocean Water). In these measurements, wood (C-5) and α-cellulose (C-3) reference materials from IAEA were used.

The isotopic discrimination in photosynthesis was calculated according to the model of Farquhar and Lloyd (1993): Δ<!-- ? -->13C=δ<!-- d -->13Cair−<!-- - -->δ<!-- d -->13Cplant1+δ<!-- d -->13Cplant1000$$\begin{array}{} \displaystyle \Delta^{13}C=\frac{\delta^{13}C_{air}-\delta^{13}C_{plant}}{1+\frac{\delta^{13}C_{plant}}{1000}} \end{array} $$(2.2)Δ<!-- ? -->13C=a+(b−<!-- - -->a)Ci/Ca$$\begin{array}{} \Delta ^{13}\rm{C=a+(b-a)Ci/Ca} \end{array} $$(2.3)

where C_i is intercellular CO₂ concentration, C_a is ambient CO₂ concentration, a (ca. 4.4‰) is the discrimination against ¹³CO₂ during CO₂ diffusion through stomata, b (ca. 27‰) is the discrimination associated with carboxylation.

The average width of the radial increment of the studied pine populations ranged from 1.31 mm to 2.20 mm (Table 1). Tree-ring width chronologies, apart from their short-term variance are characterised by a clear long-term variance (Fig. 4). Chronologies of sensitivity display a strongly reduced long-term variance while the short-term variance is highlighted. The average sensitivity of pines for the period of 1950–2012 was included in a narrow range from 0.214 to 0.281, while the r_bt index was between 0.399 and 0.521 (Table 1). On the other hand, in all cases the EPS considerably exceeded the minimum value of 0.85. These results indicate a relatively high sensitivity of the pines, the strength of the climatic signal and representativeness of site chronology. The similarity of year-to-year changes of the tree-ring width of 6 pine populations was relatively high. Tree-ring width chronologies also had a similar long-term course.

Fig. 4

Table 1

Statistical characteristics of the site chronologies for the 1951–2012 period. TRW – mean tree-ring width; TRI – mean tree-ring index; MS – mean sensitivity; r_bt – the mean correlation of standardized series of trees – the signal strength of the chronology, EPS – the expressed population signal is a statistic for examining the common variability in a chronology.

In the 1970s the pines strongly reduced their radial increments (Fig. 4). This was a result of the increase in industrial pollution at this time (Fig. 1). The reduction in tree rings radial growth in 2003 and 2006 were due to drought events

It is interesting that the year-to-year sensitivity of trees remained at a similar level throughout the analysed period (Fig. 4). The changes of tree-ring widths from year to year are the result of the impact of the climate on trees. Therefore, it can be assumed that the climate signal contained in the chronologies of pine populations was at a similar level in the analysed period.

Short-term changes of radial increments of the pine populations differed. This diversity is indicated by the location of the site sensitivity chronologies in relation to component loadings of PC1 and PC2 (Fig. 5). Sensitivity chronologies are strongly differentiated mainly by PC2. The A chronology positively correlates with PC2, while the correlation is negative in the case of S (Fig. 5). Other chronologies (B, C, N and T) show very poor correlation with PC2. They form a compact group and correlate strongly with PC1, while the chronology of populations A and S correlate poorly with PC1 (Fig. 5). The PC1 accounts for 78% of the variance among 6 sensitivity chronologies and the PC2 accounts for 10%. PC1 was most effective for explaining the variance of the radial growth of the pines.

Fig. 5

Climate determines the short-term variance of incremental reaction of trees. This kind of variance is illustrated by sensitivity chronologies. It can therefore be assumed that the two main components – PC1 and PC2 – depict the climate factors that had a significant impact on the variance of radial increments of pines. This is confirmed by the results of cluster analysis. It was found that a series of correlation coefficients for the populations B, C, N and T formed a single cluster, while the series of populations A and S differed significantly (Fig. 6).

Fig. 6

With the response function analysis, the climate features described by PC1 and PC2 were identified (Fig. 7). It turned out that PC1 correlated significantly (p<0.05) with monthly values of relative humidity and precipitation of September of the preceding year, humidity, temperature and sunshine duration of spring months and with rainfall, temperature, and sunshine duration of the summer months in the year of ring formation (Fig. 7). On the other hand, PC2 correlated significantly (p<0.05) with the humidity, sunshine duration, temperature and rainfall of spring and summer months of the current year, and in addition with the rainfall in January and temperatures of September and December of the previous year (Fig. 7).

Fig. 7

In 3 cases PC1 and PC2 were significantly correlated with the same climate parameters. This applies to relative humidity, rainfall and sunshine duration in May of the year of ring formation (Fig. 7). The impact of climatic conditions of this month on the radial increment of pines therefore requires thorough consideration. The principal component analysis showed that the climatic factors described by PC1 have a similar effect on the radial increment of 6 pine populations. In turn, climate elements described by PC2 were differentiated by the incremental rhythm in pine growth (see Fig. 5). The above results were verified by correlating these climate parameters with site sensitivity chronologies (Fig. 8). It turned out that the relation between a relative radial increment and the temperature of February, March and August were very similar in all populations (Fig. 8). These results correspond with the results obtained for PC1 (Fig. 8).

Fig. 8

The results were similar in the case of rainfall of the previous September and of current May and July. The similar sensitivity of 6 pine populations to the amount of sunshine duration in April, May and August was also discovered. Similar patterns are connected with the effect of relative humidity of the previous September, and current February, March and April (Figs. 7 and 8).

The results of response function analysis performed for PC2 were also confirmed. It turned out that the incremental sensitivity of pine populations A and S to the temperature of the previous September and December and current June and July was different. It also differed from the sensitivity of the remaining pine populations (B, C, N and T) (Fig. 8). A difference was also observed regarding the effect of precipitation of January, April, May and June of the current year on increments of the A and S populations (Fig. 8). Differences were also found in the responses of pine trees on site A and S to the sunshine of the previous September and in the period between May and July of the year of ring formation. Also, the humidity of May, July and August had a different effect on the radial increment of pine trees on both of the above positions (Fig. 8). These climatic parameters were described by the other main component which differentiated the incremental rhythm of population A and C (Fig. 7).

The impact of rainfall, sunshine duration and relative humidity in the May of the ring formation year on pine radial increments was specific. On the one hand, the conditions above have had a similar effect on the incremental reactions of pines in all populations, as monthly values of these elements significantly correlate with PC1. On the other hand, they differentiate the incremental rhythm of each population, as they significantly correlate with PC2 (see Figs. 7 and 8). This paradox is explained by the results of the site chronology correlation analysis with the climatic parameters of May. It turned out that the direction of these relations is similar in 6 populations. However, increments of the S population of pines do not show a significant relationship with the above parameters of May, and population A has the strongest relation with the above (Fig. 8). It shows the similarity of the relation, but also highly diversified strength of these relations.

Analysis of stable isotope fractionation in annual tree rings of pines, growing in the sampling site where the strongest reduction of tree ring width was noted, is the complemental analysis to dendrochronological method. The pine tree-ring isotopic chronologies constructed in this study is a local grown pattern of the partial pine population (site C). These isotopic records (Fig. 8) are the result of the response of the trees to variations of various environmental factors such as climatic factors and human activities.

δ¹⁸O values varied between 27.4‰ to 31.8‰ (Fig. 9), whereas δ¹³C ranged from −24.6‰ to −22.9‰. The δ¹³C changes in the isotopic ratio of atmospheric CO₂ due to global fossil fuels emissions ranged from 0.37‰ (in 1975) to 1.97‰ (in 2012). This correction lifts the δ¹³C values (Fig. 9). The δ¹³C values corrected for changes in the isotopic ratio of atmospheric CO₂ ranged from −23.6‰ to −21.5‰.

Fig. 9

The analysis of combined measurement of carbon and oxygen isotopes ratio (Fig. 10) show positive significant correlation between δ¹³C and δ¹⁸O in the period of time from 1991–2012. No significant δ¹⁸O-δ¹³C relationship was observed between 1975 and 1990.

Fig. 10

To describe the variation of the carbon and oxygen isotope composition of cellulose in annual tree-rings of pine caused by climate changes and human activities we used a model (Sensuła and Pazdur, 2013a, 2013b) based on multiple regressions (Fig. 11, Table 2): δ<!-- d -->=ba+∑<!-- ? -->M=AprSeptbMTTM+∑<!-- ? -->M=AprSeptbMPPM+∑<!-- ? -->M=AprSeptbMSSM+∑<!-- ? -->M=AprSeptbMHHM$$\begin{array}{} \displaystyle \delta=b_a+\sum\limits_{M=Apr}^{Sept}b_{MT}T_M+\sum\limits_{M=Apr}^{Sept}b_{MP}P_M+\\ \displaystyle\sum\limits_{M=Apr}^{Sept}b_{MS}S_M+\sum_{M=Apr}^{Sept}b_{MH}H_M \end{array} $$(3.1)

where M is the month (from April to September), b is the regression coefficient for the following variables: T (mean of the monthly maximum temperatures), P (monthly precipitation sum), S (monthly hours of sunshine duration), H (mean of the monthly relative humidity), whereas b_a corresponds to the interdependences between the monthly climate factors and other environmental changes.

Fig. 11

Table 2

The regression coefficient (Eq. 3.1) for the following variables: T (mean of the monthly maximum temperatures), P (monthly total precipitation), S (monthly hours of sunshine duration), H (mean of the monthly relative humidity). b_i – multiple regression parameters. The test of significance gives a p-value (n=38).

The value of the correlation coefficient between the measured and modelled δ¹³C in α-cellulose is equal to 0.93, whereas the value of the correlation coefficient between the measured and modelled δ¹⁸O in α-cellulose is equal to 0.78.

Diffuse air pollution (carbon dioxide) caused the variation in the ratio of water used in plant metabolism to water lost by the plant through transpiration (WUE). In the period of time from 1975 and 2012, according to NASA (2016) the global concentration of mid-tropospheric carbon dioxide ranged from ca. 330 ppm to ca. 393 ppm whereas the global surface temperature, relative to 1951–1980 average temperatures, ranged from −0.01°C to 0.76°C. During this period, the water use efficiency values increases from 94 to 122 µmol/mol (Fig. 12). It has been observed that with increase of CO₂ emission also WUE increases. Water use efficiency might be strongly correlated with variability of the surface temperature (r=0.81)

Fig. 12

Site chronologies of radial increment sizes represent incremental patterns of pines from partial populations. They illustrate the specific response of pine trees to various environmental factors, including the climate factor. It should be emphasized that in the years 1951–2012 the size of radial increments of the studied pine population changed. In the 1970s pines strongly reduced their radial increment. The reason for the reduction was a strong increase in industrial pollution. Despite this fact, pines in the area preserved their high year-to-year sensitivity to short-term impulses from the environment. It turned out, however, that the year-to-year rhythm of radial increments in particular populations was different. The reason was their different sensitivity to particular elements of the climate. In particular, two pine populations (S and A) were different in this respect – both, from each other and from the other four populations. They are located away from the main flow of air masses carrying pollutants from local factories. It is therefore assumed that the differences in the relations between climate and radial increments were the result of the varying pressure of pollution on the studied populations of trees. However, the influence of other factors, such as unrecognized differences in habitat conditions and the age of the trees cannot be excluded. Site A pines were in fact younger than others by approx. 20 years.

The differences in sensitivity of pines from A and S populations to the climatic elements described by PC2 were confirmed by the analysis of site chronology correlations. The Principal component analysis (PCA) is therefore an effective method of recognizing the similarities and differences in the sensitivity of trees to the climatic factor.

Interpretation of the modifying effect of pollutants on the relation between climate and tree increments is difficult. The influence of pollutants depends on the distance of trees from the emitters, the position of sites in relation to the direction of wind carrying pollutants, the age of trees and habitat conditions in which trees grow (Carrer and Urbinati, 2004; Yu et al., 2008; Friedrichs et al., 2009; Dauskane et al., 2011; Wilczyński, 2013; Sensuła et al., 2015a, 2015b).

A significant influence of the climatic conditions of the current and previous year on pine radial increments was observed. This is confirmed by numerous dendroclimatic studies on Scots pine (Vaganov, 1990; Lührte, 1991; Richter et al., 1991; Lindholm et al., 1996; Wilczyński and Skrzyszewski, 2003; Wilczyński, 2003, 2013; Pensa et al., 2005; Tuovinen, 2005; Juknys et al., 2014; Helama et al., 2013).

A strong positive correlation of site chronologies with the first main component indicates that pines showed similar sensitivity to climatic influences of PC1. Their role in shaping thickness increment of pines was independent of other elements of the environment. The results indicate that a cool, humid and rainy September meant that all the pine populations increased their increments in the following year. High temperatures and low air humidity in the autumn has a positive effect on the flowering of female flowers in the following year (Andersson, 1965; Fober, 1976). This has a negative impact on incremental growth of trees (Eis et al., 1965; Chałupka et al., 1976). On the other hand, cloudy, humid weather with plenty of rainfall has a beneficial effect on the number of buds of vegetative organs – needles and shoots (Hejnowicz, 1982). Therefore, the surface of assimilation and production of auxins increases, which speeds cambial divisions and growth of trees. It was also found that hot, short and sunny winters and dry air in this season positively influence the condition and incremental ability of 6 populations of Scots pine. Such conditions gave an earlier start to physiological processes that led to earlier and faster divisions of the cambium (Schober, 1951; Ermich, 1959; Wodzicki and Zajączkowski, 1983). A similar positive role was played by a dry, sunny and warm spring. These relations are confirmed in other dendroclimatic studies of the pine (Richter et al., 1991; Wilczyński and Skrzyszewski, 2003; Wilczyński, 2003; Juknys et al., 2014). The positive impact on the creation of the pine wood cells was also exerted by high rainfall in July of the year when the ring was formed, as they caused the extension of the period of intensive divisions of vascular cambium and the formation of large diameter cells. Low rainfall in summer is often a factor limiting the incremental growth of Scots pine basically throughout all of its range (Lührte, 1991; Lindholm et al., 1997; Irvine et al., 1998; Cinnirella et al., 2002; Wilczyński and Skrzyszewski, 2003; Pensa et al., 2005; Tuovinen, 2005; Pilcher and Oberhuber, 2007; Piovesan et al., 2008; Gruber et al., 2010; Juknys et al., 2014). Drought is a factor which often induces variability in growth patterns and mortality of Scots pine especially in dry areas (Bigler et al., 2006, Hereş et al., 2012, Herguido et al., 2016, Marqués et al., 2016). Heavy rainfall during the growing season and, consequently, increased air humidity enhances the flow of water into the cells increasing the pressure on the cell wall, thus stretching these and causing cambium cell growth and it also provides the substrates for the process of photosynthesis (Major and Johnsen, 2001). In turn, the deficit of water within this period reduces the turgor of the cell, decreasing the dividing activity of the vascular cambium (Pichler and Oberhuber, 2007). It also happens that excess of water harms trees, but mostly those on very humid sites (Cedro and Lamentowicz, 2011; Helama et al., 2013; Wilczyński, 2013).

The determination of properties of tree-rings is crucial for many applications in the investigation of local and global environmental changes. Since the beginning of the 20^th century, there has been much discussion about how external environmental factors, including climate changes and anthropogenic effects affect the physiological processes that control tree growth (Schweingruber, 1996; DeVries et al., 2000; McCarroll and Loader, 2004; Pazdur et al., 2007, 2013; Sensuła et al., 2015a, 2015b; Sensuła, 2015). Through photosynthesis, plants convert CO₂ and H₂O to saccharides (C₆H₁₂O₆), using light, and release oxygen to the atmosphere. The carbon of annual tree-ring has its origin in the CO₂ of air, whereas the oxygen has its origin in the soil water and thus precipitation. The primary value of the isotopic records in tree-rings is not simply as samples of ancient air or water, but as sensitive bio-indicators of the way that the components of air and water have been changed by the trees in response to the environments in which they grown (Craig, 1954; Ehrelinger and Vogel, 1993; McCarroll and Loader, 2004; Savard, 2010; Sensuła, 2015).

The carbon isotopic composition (δ¹³C) of trees has been influenced by carbon isotopic composition of atmospheric CO₂, diffusion of CO₂ through stomata, and enzymatic discrimination during the irreversible step of CO₂ fixation (Roden et al., 2000). The isotopic composition of plants varied due to emission of atmospheric pollution (Savard, 2010), such as for example sulphur dioxide (Rinne et al., 2010), carbon dioxide (Craig, 1954; McCarroll and Loader, 2004; Pazdur et al., 2007, 2013; Sensuła et al., 2011a, 2011b; Sensuła and Pazdur, 2013a, 2013b; Sensuła, 2015).

Trees grown at the higher level of CO₂ concentration had a more negative δ¹³C than trees grown at the lower concentration. In pine, CO₂ usually limits photosynthesis and, thus, an increase in CO₂ results in greater photosynthetic rates. Raw δ¹³C data can be corrected to a preindustrial atmospheric δ¹³C_cor base value of 6.4% using the data from McCarroll and Loader (2004), due to decreasing δ¹³C in the air and the biosphere is associated with the increasing of anthropogenic CO₂ in the atmosphere (Craig, 1954; Farquhar and Lloyd, 1993; Field et al., 1995; McCarroll and Loader, 2004; Pazdur et al., 2007, 2013; Keeling et al., 2010; Sensuła and Pazdur 2013a, 2013b; Battipaglia et al., 2014; Saurer et al., 2014). The analysis of the influence of pollution on stable isotopes composition will be a theme of the future study. Carbon dioxide (CO₂) is an important heat-trapping (greenhouse) gas, which is released through human activities such as deforestation and burning fossil fuels, as well as natural processes such as respiration and volcanic eruptions. According to NASA (2016), and IPCC (2005), the extent of climate change effects on individual regions will vary over time and with the ability of different societal and environmental systems to mitigate or adapt to change. The IPCC predicts that increases in global mean temperature of less than 1 to 3 degrees Celsius above 1990 levels will produce beneficial impacts in some regions and harmful ones in others. Net annual costs will increase over time as global temperatures increase.

Researchers have used correlation analyses to determine which environmental parameters (precipitation, sunshine, humidity, and temperature) might be recorded in the isotopic composition of cellulose extracted from the annual growth rings (Schiegl, 1974; Gray and Thompson, 1976; Epstein and Yapp, 1977; Burk and Stuiver, 1981).

According to the scientific literature, the carbon isotopic composition in plant can vary with water stress and solar radiation and can be correlated with amount of rainfall, vapour pressure deficit, canopy position and hydraulic conductivity associated with tree height (Dongmann et al., 1974; Ehleringer, 1990; Ehrelinger and Vogel, 1993; Comstock and Ehleringer, 1992; Buchmann et al., 2002; Yoder et al., 1994; Barbour et al., 2002), whereas the oxygen isotopic composition of trees can be influenced by the oxygen isotopic composition of meteoric water and atmospheric vapour, atmospheric humidity and vapour pressure deficit (Farquhar and Lloyd, 1993; Roden et al., 2000). The oxygen isotopic composition of plant has been correlated with air temperature, relative humidity, transpiration rates and water balance, presumably all through contributions of source water (the xylem water in suberized stems) and leaf water enrichment (Ehrelinger and Vogel, 1993). In our studies, the impact of weather conditions on the isotopic concentration in pine was observed. The most significant climate factors (p<0.05, where p is the significance level) influencing δ¹³C is the monthly humidity in September (p=0.026), whereas the most significant climate factors influencing δ¹³C_cor are the monthly humidity in June (p=0.009) and September (p=0.001), sunshine in June (p=0.003), July (p=0.045) and September (p=0.013), whereas the most significant climate factors influencing δ¹⁸O in the investigated area is the mean of the May monthly maximum temperatures (p=0.043). The climatic signal record in trees stable isotopes chronologies may be masked by air contamination. Previous studies have shown weakened climate signal in δ¹³C and δ¹⁸O of α-cellulose of tree ring cellulose as well as deviant trends between δ¹³C and δ¹⁸O series caused by pollution during the 20^th century (Rinne et al., 2010, Boettger et al., 2014, Sensuła, 2016b).

According to models (Farquhar and Lloyd, 1993; Scheidegger et al., 2000) a positive correlation of the δ¹⁸O and δ¹³C is predicted when Δ¹³C is driven by a strong stomatal reaction, whereas A_max (photosynthetic capacity) is relatively unaffected. Saurer et al. (1997) found a positive correlation between δ¹⁸O and δ¹³C of stem organic matter for pine trees growing in habitats with moist and dry soil conditions. The slopes in the δ¹⁸O-δ¹³C plots were found to be species-dependent and were interpreted as due to differing sensitivity to the moisture conditions. Generally, a positive correlation between oxygen and carbon isotopes might be expected in situations when water is not limiting, because there is no need to reduce water loss, which allows the stomata to operate over a wide range (Scheidegger et al., 2000).

A negative δ¹⁸O-δ¹³C relationship could be regarded as an indication that plants have stomata with a limited operational range (Scheidegger et al., 2000). According to Saurer and Siegwolf (2007) the stronger response of photosynthetic capacity (Amax) indicate that some species were able to enhance biomass accumulation due to increasing CO₂ during the last decade, whereas other species responded more strongly with reduced stomatal conductance and less transpiration and water loss.

Spatial variability and temporal trends in water-use efficiency of European forests up until 2000AD has been studied in several tree species. Experimental results show that plants are able to increase their water-use efficiency (WUE) as CO₂ levels rise (Ehlelinger et al., 1993; Morison, 1993; Field et al., 1995; Gagen et al., 2011; Saurer et al., 2014; Sensuła, 2015; 2016a). Increased atmospheric CO₂ concentration may stimulate plant growth, indirectly through reduced plant water consumption and hence slower soil moisture depletion, and directly through enhanced photosynthesis (Morgan et al., 2004). According to NASA (2016) the variability in the global surface temperature might be due to increase of global CO2 emissions. According to NASA (2016) the land surface temperature is described as a measurement of how hot the land is to the touch. It differs from air temperature because land heats and cools more quickly than air. Most recent literature (for example, Stips et al., 2016) confirm that the total greenhouse gases (GHG), are the main drivers of the changing global surface air temperature. It has been observed that the elevated CO₂ increases intrinsic water use efficiency (WUE) of forests, but the magnitude of this effect and its interaction with climate is still poorly understood. The analysis of the influence of variability of the global surface temperature due to increase of global CO₂ anthropogenic emissions on WUE will be a subject of future analyses.