Next Article in Journal
Decreased Soil Microbial Biomass and Changed Microbial Community Composition following a Defoliation Event by the Forest Tent Caterpillar
Previous Article in Journal
Adaptive Trait Variation in Seedlings of Rare Endemic Mexican Spruce Provenances under Nursery Conditions
 
 
Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
Journals
Forests
Volume 14
Issue 4
10.3390/f14040791
forests-logo
Submit to this Journal Review for this Journal Propose a Special Issue
► Article Menu

Article Menu

share Share announcement Help format_quote Cite question_answer Discuss in SciProfiles thumb_up
...
Endorse textsms
...
Comment
first_page
settings
Order Article Reprints
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

Direct and Carry-Over Effects of Temperature Drive Height Increment of Scots Pine in the North-Eastern Baltic Sea Region

by
Roberts Matisons
1,*,
Sandra Metslaid
2,
Maris Hordo
2,
Regino Kask
2,
Ahto Kangur
2,
Hannu Salminen
3 and
Āris Jansons
1
1
Latvian State Forest Research Institute ‘Silava’, 111 Rigas Str., LV-2169 Salaspils, Latvia
2
Institute of Forestry and Engineering, Estonian University of Life Sciences, Kreutzwaldi 5, 51006 Tartu, Estonia
3
Finnish Institute of Natural Resources LUKE, Ounasjoentie 6, 96200 Rovaniemi, Finland
*
Author to whom correspondence should be addressed.
Forests 2023, 14(4), 791; https://doi.org/10.3390/f14040791
Submission received: 1 March 2023 / Revised: 31 March 2023 / Accepted: 6 April 2023 / Published: 12 April 2023
(This article belongs to the Section Forest Meteorology and Climate Change)
Browse Figures
Versions Notes

Abstract

:
In the eastern Baltic region, the abundance of Scots pine (Pinus sylvestris L.) has been predicted to shift due to changes in height growth and competitiveness. Under such conditions, the relationships between tree growth and meteorological/climatic conditions can provide valuable information on the ecological plasticity and adaptability of local populations. Regarding the climatic dependence of productivity and competitiveness, height increment is an informative proxy, although data gathering is laborious. In this study, meteorological sensitivity of the height increment of Scots pine across the climatic gradient of north-eastern Baltic region was assessed by the means of timeseries deconstruction and generalized additive regression. Previously published time series from sites in Latvia, Estonia, and Finland were reanalysed. The local linear weather–growth relationships showed gradual changes in sensitivity to low summer temperature to conditions related to water availability in summer in Finland and the Baltics, respectively. The regional generalization of responses indicated the prevalence of temporary and spatially stationary effects of temperature in winter and summer, which had a complex influence (direct and carry-over effects). The prevailing linearity of the regional responses suggested that, regarding height growth, the studied populations appeared adapted to a wider climatic gradient than the studied one.

1. Introduction

In the eastern Baltic Sea region, conifers are predicted to gradually surrender to deciduous trees, thus shifting forest composition [1], as the adaptability of the native conifers appears to be lagging behind the accelerating changes in environment [2,3]. The effects of climate change on forest growth and productivity are disproportional [4,5,6] and vary spatially [7,8,9], thus maintaining uncertainties in the projections [10]. Accordingly, comprehensive information on the environmental effects on tree growth is still essential for sustainable adaptive forest management [11,12].
The relationships between increment and environmental fluctuations are more informative regarding growth performance under the shifting environment compared to the dimensions of trees, which are the cumulative product of past conditions [11,13,14,15]. Due to convenience, growth sensitivity has been commonly assessed based on the radial increment [9,11,12]. Height increment, which is a better indicator of productivity and competitiveness, while providing complementary information on growth sensitivity [16,17,18,19,20], has still received little attention due to the laborious measurements [21,22]. Hence, analysis of the variation of height increment and its components can provide deeper insight into the environmental limitations of tree growth [7,14,23,24].
In the eastern Baltic Sea region, Scots pine (Pinus sylvestris L.) is a widespread and economically important species projected to decrease abundance (e.g., up to 40% over 21st century in the Baltics) [1,25]. Such a shift can be amplified by local adaptation, which restricts ecological plasticity despite intensive gene flow [2,3,6,26,27]. The decline of native genotypes, which are not able to benefit from a warmer climate, is largely driven by competition [20,25,28]. Under such conditions, information about the environmental sensitivity of height increment, and hence productivity and competitiveness of trees [15,18,24], can aid adaptive management [29]. This is particular for Scots pine due to the pronounced weather sensitivity of the height increment [15,17,18,21,24,30].
The formation of the height increment of Scots pine is a two-year process depending on the number and elongation of growth initials formed at the end of the preceding growing season [31,32]. Hence, the height increment is affected by conditions over a longer period compared to tree-rings [15,18,33], particularly as the formation of growth initials can be affected by the preceding conditions (i.e., experience carry-over effects [15,17,30]. Accordingly, the climatic controls of height increment appear to be complex, as the effects of meteorological conditions interact [4,5,8,17,24,34]; for instance, temperature modulates the effects of water deficit [2,35,36].
Under the cold climate of northern Finland, the height increment of Scots pine has been driven (linearly depended on) by temperature in the preceding summer [15]. In those stands, the current summer temperature and precipitation showed local significance [15] alarming of water shortages [35,36]. Under the warmer climate in Latvia, the height increment lacked a strict weather limitation, as the sets of weather–growth correlations differed between the coastal and inland sites [17,30]. Nevertheless, the conditions in the preceding year were the main (though nonstationary) drivers of increment; the conditions in the current year had secondary effects [17,30]. Accordingly, the weather–growth correlations suggested complex regional meteorological controls over increment [8,15,17,24].
Across the environmental gradient, ecological responses are generally bell-shaped [9,37], while the linear responses can be estimated under a limited part of the gradient [28,37,38]. Unfortunately, under rapid shifts in environmental gradients, the local linear relationships, which are widely used for projections, appear to be biased and nonstationary [9,39]. Accordingly, the assessment of ecologically realistic responses across the environmental gradient scalable with the distribution of the population(s) is crucial for reliable projections [6,9,15,38]. Furthermore, the nonlinearity of weather–growth responses implies their dependence on the local climate, highlighting the disproportional effects of climate change on forest composition and productivity [4,5,6,38]. Nonlinear responses can also be attributed to the nonstationarity of linear relationships, which emerge as the climatic gradient shifts [9,39].
The aim of the study was to generalize the responses of the height increment of Scots pine to weather conditions and to assess their plasticity under the climatic gradient of the north-eastern Baltic Sea region. The assessment of the main weather-related drivers of height increment and the evaluation of the stationarity of the local linear relationships were considered as the subordinate objectives. Considering the location of the study region in the temperate and boreal climatic zones, we hypothesized that both the temperature and precipitation were the regional drivers of increment. We also assumed that the responses to summer temperature and precipitation would be arc-shaped, indicating the presence of optimum conditions.

2. Material and Methods

2.1. Dataset

The generalization of the responses of the height increment of Scots pine to meteorological conditions was based on the previously published height increment time series from the north-eastern Baltic Sea region [15,17,21]. In total, sample data representing six closed canopy stands in Northern Finland, Estonia, and Latvia were studied (Figure 1). To the authors’ knowledge, the studied dataset includes all published time series of height increments of mature Scots pine from the region. Data for the stands north from Rovaniemi were not analysed, as they represent open-canopy forests [15].
The studied stands differed by mean age, and most of them were growing on mesotrophic mineral soil (Table 1). The oldest stands were in Latvia and Estonia, where trees were larger and taller accordingly. Generally, site conditions were common for conventionally commercial stands of Scots pine within the studied localities, considering the regional differences of the climatic and edaphic backgrounds. Stands in Estonia were sown, while the others have regenerated naturally.
The climatic conditions (during the past five decades) at the sampled stands were generally comparable in the Baltic states except from the Finnish one (FIN_1), which was located in a different climatic zone; hence, its climate was considerably colder (particularly winter) and drier (Table 2). Hence, it must be admitted that the studied sites rather discretely represent the regional gradients. Still, the studied stands in the Baltic states represented local thermal and continentality gradients from coastal to inland according to distance from the Baltic Sea in the south-western direction (according to the dominant westerlies). The annual precipitation regime, however, was comparable, with half of the annual precipitation falling during the summer months. The climatic changes for the studied sites were similar, and they were expressed as warming during the dormancy period causing an extension of the growing period [40]. Concomitantly, summer temperature and precipitation regimes have become more heterogeneous, particularly as the warmer precipitation-free periods have been extending [40].
Dominant or co-dominant visually healthy and straight trees were selected for sampling within each stand. In general, these trees represented the size distribution of the canopy trees of the stands [15,17,21]. The number of sampled trees ranged from 10–20 per stand; the sampling was conducted during 2003–2014. The selected trees were felled and their stems were cut longitudinally uncovering the pith. Along the pith, whorls were identified and annual borders of consecutive height increments were identified (by colouring and changes in width of the pith) and marked. The dating of the borders of height increments was cross-checked with tree-ring data from the ends of the logs, thus revealing any possible leader changes (loss of the apical shoot), which could be hidden or difficult to identify. Any obscure parts of stems (with indistinctive height increment borders of tree-rings) were cut and processed in the laboratory, thus warranting the dating of increment. Height increment, as the distance between the borders of height increments on the pith of the stem, was measured using a measurement tape with an accuracy of a millimetre. Although the process was time-consuming, time series of highly accurate measurements were obtained.
A gridded monthly climatic dataset (CRU TS4 v. 4.06) was acquired from the online repository of CRU and used for the calculation of meteorological variables to be tested as predictors of height increment [41]. The climatic variables used in the analysis were mean monthly temperatures, precipitation sums, and standardized precipitation evapotranspiration index [42] calculated considering three-month window to describe the environmental water deficit. Considering that the formation of the height increment of Scots pine is a two-year process [33,43] and that meteorological conditions can have carry-over effects on increment [4,5,17,32], the meteorological variables were arranged according to a time window from June two years prior (bprev.) to the formation of increment to August in the year of increment. Such an approach has been widely used for the analysis of increment using the time series deconstruction techniques [9,11,15,17].

2.2. Data Analysis

Considering the relevance of exact dating of increment for the assessment of weather–growth relationships [44,45], graphical and statistical cross-dating using COFECHA software [44] was performed for the height increment time series within and among the studied stands. Special attention was paid to event years (abrupt improvements/suppressions of increment). Corrections in the time series were introduced if necessary. Series showing low agreement (concordance) with others within a stand without the possibility of obvious and reliable correction were omitted from further analysis. For each site dataset, a number of dendrochronology statistics were calculated to evaluate the strength of the common growth signal between the trees and to assess chronology reliability. The following statistics commonly used in radial increment analysis was calculated: first order autocorrelation, mean sensitivity, Gini coefficient, mean Gleichläufigkeit, signal–noise ratio, and expressed population signal (EPS) coefficients [46,47]. These statistics were calculated for time series detrended by a flexible cubic spline with a wavelength of 40 years and a 50% cut-off frequency to avoid possible bias related to disturbances [39,45].
To assess the effects of weather conditions on the height increment of Scots pine, the trends related to ageing and disturbances (local and site specific) were omitted from the time series of height increment [45,46]. Double detrending by the modified negative exponential curve and flexible cubic spline with the rigidity of 40 years and 50% cut-off frequency was used to highlight the inter-annual (high-frequency) variation of increment. This approach was used to stabilize inter-annual variation, as stands differed in age and management history. To represent the height increment of stands, the detrended individual time series of trees were prewhitened by the first order autoregressive model and averaged into the site residual chronologies using the biweight robust mean [46]. Stand–level linear relationships between a high-frequency variation of height increment represented by the residual chronologies and the meteorological variables were identified by a bootstrapped Pearson’s correlation analysis [48] for the reference period of 1955–2003 (common for most datasets). Non-parametric percentile interval bootstrapping with replication was used to estimate the statistical significance of the correlation coefficients at α = 0.05.
To assess regional weather and climatic drivers of height increment of Scots pine accounting for possible nonlinearity of the growth responses [9,37], a generalized additive mixed model (GAMM) was used [49]. Such models, able to estimate nonlinear splines, have been sufficiently used for the wide-scale analysis of heterogenetic ecological data representing considerable spatiotemporal gradients [4,5,6,37]. Accordingly, such a statistical method aids the assessment of more reasonable, stationary, and ecologically realistic responses [6,9].
The height increment indices of the residual chronologies for the reference period (1955–2003) were used as the response variable, while different combinations of the meteorological variables (as smooth terms) were tested as the fixed predictors (effects). The arbitrary selection of the fixed predictors was based on the correlation analysis results and Akaike’s information criterion, as well as the biological meaningfulness. The aggregated seasonal weather variables (e.g., mean temperature for January–March) were also tested if there were indications from the correlation analysis for such relationships. Meteorological variables included in the refined model were tested for collinearity by calculating the variance inflation factor (VIF), variables with VIF > 5.0 were excluded from the model. To account for spatial and temporal variability of responses and dependencies in the studied height increment datasets, as well as differences in the growing conditions, the year of increment formation and stand (site) were used as crossed random effects (random intercepts).
To eliminate the possibility of overfitting, the basis dimensions of smoothing splines for the fixed effects (meteorological variables) were restricted to four, implying that responses with up to three inflection points (as in bell-shaped responses) could be estimated. The estimation of the smoothing splines (using isotropic penalized thin plate regression without a point constraint) was conducted according to the generalized cross-validation procedure; regression splines with shrinkage were fitted [49]. For conventional cross-validation of the refined model, 25%–30% (randomly selected) of the cross-dated measurement time series were excluded from the model calibration and used for verification (all chronologies were calculated as described above). Root mean square error was used as an indicator for validation of the responses. The model was fit using the restricted maximum likelihood approach [49]. Model residuals were inspected for compliance with the statistical assumption using diagnostic plots; the performance of models was judged by relative, absolute, and root mean square errors, as well as R-squared values, etc. Data analysis was conducted in R software v. 4.2.2 [50] using packages ‘dplR’ [46], ‘treeclim’ [48], and ‘mgcv’ [49]. TURPINAAT

3. Results

After the statistical and graphical cross-dating, from 65% to 100% of the measured time series of height increment per site successfully passed quality checking, implying correct dating and reasonable similarity of inter-annual variation. Although regional data have been analysed, the statistics of the cross-dated datasets indicative of environmental signals generally lacked explicit latitudinal trends. The mean height increment calculated for the full increment series span site ranged from 28.5 to 58.0 cm, being greater in younger pine sites from Estonia. The standard deviation of the increment was similar for the trees in the Baltic states, but twice as small in the northernmost site (Table 3). The annual variation of height increment was little affected by preceding growth, as indicated by the weak to intermediate first order autocorrelation calculated for the detrended series. The inter-annual variation was rather limited (being intermediate at most), as shown by low values of the mean sensitivity and Gini coefficients, implying restricted environmental effects on primary growth. The mean interseries correlation indicating local forcing of growth, however, differed among the stands ranging from low to intermediately high in EST_1 and EST_3, respectively. The above-average synchronicity of the datasets suggested that the inter-annual variation of the increment among trees was of a similar phase yet differed in extent. Nevertheless, the studied datasets contained sufficient environmental signal(s), as indicated by the values of EPS ≥ 0.85 (cf. [47]), although the strength of the environmental signal was intermediate, as indicated by the signal-to-noise ratios.
The cross-dated time series of the height increment from Latvia and Estonia showed a similar trend with low values during the first ten years followed by rapid growth and a gradual decrease afterwards (Figure 2). Under the colder climate of Northern Finland, the height increment stabilized after the period of juvenile slower growth. The event years (abrupt increases/decreases within the majority of trees) in the height increment time series were not expressed; hence, the inter-annual variation was moderate, as indicated by mean sensitivity. Still, a common abrupt decrease was observed in Latvia in 1930. The individuality of the inter-annual variation pattern of height increment of the studied trees increased with age, particularly when exceeding 50 years of age. The replication of the datasets was high throughout the reference periods (Figure 3). The developed residual chronologies of stands showed a gradual geographic shift in the variation pattern between the coastal and the northernmost stands. The correlation among them, however, was low (0.19). Double detrending was sufficient to normalize the inter-annual variation of the height increment throughout the reference period, allowing the extraction of an environmental signal.
The assessment of local linear weather–growth relationships indicated that 29 of the 81 meteorological variables tested showed significant correlations with the residual chronologies of the height increment in at least one stand (Figure 4). The strength of the significant correlations was weak-to-moderate, ranging 0.31–0.70. The variables related to the conditions one and two years before the formation of the increment indicated legacy effects, yet those dating with a growing period revealed the direct effects of meteorological conditions. Significant correlations with both temperature- and precipitation-related variables of different timing indicated complex meteorological controls over the height growth of Scots pine.
A gradual latitudinal shift in the significant weather–growth correlations was evident, as those with the variables related to water availability in summer intensified southwards and as the continentality increased (Figure 4). In Northern Finland, the weather–growth correlations with temperature and precipitation indicated a clear limitation of increment by cool summers in the preceding year. A complex of correlations was explicit under warmer climates in sites in Latvia and Estonia, indicating the effects of temperature and precipitation in preceding summer and autumn, as well as winter temperature. The correlations with meteorological variables of the year before previous showed weaker correlations with increment, and those related to autumn were negative. Still, in the older stand in Estonia (EST_1), the height increment only showed weak linear relationships with SPEI in summer/autumn of current, as well as two years before growth. Nevertheless, in the Baltics, the correlation with temperature in the previous summer/autumn showed a local gradient, as the significant variables tended to date later under more continental climate. The winter temperature in the year of increment showed positive correlation in the coastal site in Latvia, while the direct effects of summer water availability were indicated by correlation with June SPEI in a site in Estonia. Additionally, winter precipitation related variables showed positive correlations.
The overall performance of the model generalizing regional weather–growth relationships can be considered high for heterogenetic ecological data, with the fixed effects able to explain ca. 38% of the variation in the data, as indicated by the marginal R2 (Table 4). Furthermore, the conditional R2 was substantially higher, indicating explicit spatiotemporal specifics of weather–growth responses represented by the random effects. Among those, the year had considerably higher variance compared to the site, indicating that temporal rather than spatial variability of responses prevailed, suggesting regional coherence and robustness. The temporal variability could likely be related to interactions of the variables, which, however, were not tested considering the limited scope of the study. The temporal autocorrelation of the responses of height increment to weather conditions was low, likely as the responses were prewhitened. The implication of generalized cross-validation and restriction of the basis dimension of smoothing splines were sufficient to avoid overfitting, as indicated by similar errors resulting from calibration and verification of the regional model, thus warranting the relevance of the estimated responses.
The generalization of regional weather–growth responses estimated a set of eight meteorological variables as the main regional drivers of the height increment of Scots pine, two of those were derivatives of the monthly means (Table 4). These variables represented both the direct and carry-over effects of meteorological conditions. The estimated regional responses indicated the prevailing sensitivity of the increment to thermal conditions, as only a single variable related to precipitation had a significant effect. Although the dataset representing regional climatic gradient was analysed, most of the regional responses were linear (effective degrees of freedom being one; Table 4), showing that the studied gradient has been rather short. Nevertheless, two variables were estimated with “quadratic” responses with one inflection point (effective degrees of freedom > 2.0) and one variable with a near-linear response.
Among the significant meteorological variables, the strongest effect on increment (highest F-value) was estimated for temperature in the preceding July, which showed a linear positive relationship with height increment (Figure 5C), as also suggested by correlation analysis (Figure 4). The second strongest effect was estimated for temperature in current August, which represents the very end of growing period and had a linear negative response of height increment (Figure 5G). Similarly, a negative linear, though weaker effect was estimated for August precipitation (Figure 5H). A direct influence of meteorological conditions on height increment was supported by the significant effect of the minimum mean temperature in the June–July period, to which increment responded positively (Figure 5F). However, this effect was estimated with the lowest F-value among the significant ones (Table 4).
The temperature in the previous October, which showed the strongest correlations under warmer climate in Latvia (Figure 4), was estimated with the third strongest effect (Table 4). The response showed a threshold of ca. 5.0 °C, exceeding which, temperature had a positive effect on increment (Figure 5D). In contrast, below-threshold temperature was estimated with a slight, though constant negative increment. A nonlinear response with a threshold value of ca. 15.0 °C was estimated for temperature in the previous June (Figure 5B), which had an intermediate effect on increment (Table 4). The response showed that temperatures below the threshold had a positive effect on increment, while that above lacked effect. Surprisingly for trees growing in a cold climate, the temperature in winter prior to the growing season (January and February) had a negative linear, although a rather weak regional effect (Figure 5E; Table 4). The extended carry-over effects of meteorological conditions at the regional scale were indicated by the significant yet intermediate effect of August temperature in the year before the previous (Table 4), which caused a near-linear negative response of increment (Figure 5A). The response became slightly steeper if the temperature exceeded ca. 15.0 °C.

4. Discussion

The height growth of trees within a forest stand is a dynamic and individual process affected by changing competition [16,17,20], as well as prevailing meteorological conditions [15,17], challenging the reconstruction of height increments [19,21,22]. Nevertheless, the cross-dated time series of height increment contained reasonable common environment signals, as indicated by the metrics of the datasets (e.g., EPS > 0.85; Table 3), implying their sufficiency for local assessment of growth responses [9,22,47]. The strict (double) detrending highlighted the high-frequency (inter-annual) variation of height increment (Figure 3), and hence, the effects of meteorological conditions on trees differing by age and management history [38,45]. The autocorrelation in the detrended height increment (Table 3), however, was lower compared to that of the radial increment (cf. [4,5]), implying lesser dependence on previous year’s growth [51]. Still, the opposite has been observed in southern Finland [22]. The mean sensitivity and Gini coefficients of the double-detrended datasets, were low to intermediate (cf. [4,5]), suggesting moderate inter-annual variation of increment, likely due to interacting effects of environmental fluctuations under non-marginal conditions [8,39,52].
The local specifics of latitudinally shifting weather–growth correlations (Figure 4), indicated effects of site conditions [8] and/or local adaptation [6,26,27]. Nevertheless, the estimated regional weather–growth responses showed a low effect of stand (random variance, Table 4), suggesting their spatial stationarity [6]. This was also supported by the distance-depending similarity visible among the chronologies (Figure 3), which indicated a gradual shift of the inter-annual variation of height increment. Low effect of stand (Table 4) might also be partially related to the spatial heterogeneity of the dataset, discretely representing the regional climatic gradient.
Common sensitivity to principal environmental drivers of growth in non-marginal parts of species distribution has been considered as an adaptation maximizing competitiveness and survival [6,52]. The variance related to year of increment was substantially higher (Table 4), implying inter-annual variability and, hence, the plasticity of responses, likely as the climatic gradient was shifting [39] and/or meteorological conditions interacted [6,8,34], enhancing their influence [2]. The local differences in weather–growth correlations among the adjacent stands in Estonia (Figure 4), however, might be attributed to differences in tree age [9,22,24]. The increasing individuality of height increment, as trees aged past 50 years (Figure 2) might also be related to growing susceptibility to xylem dysfunction [53] and, hence, micro-site [54] or reproductive effort [55].
The estimated regional responses (Table 4, Figure 5) highlighted the complexity of climatic controls over the height increment [15,17], with the mutual effects of meteorological variables driving growth in the north-eastern Baltic Sea region [5,8,15]. These responses (Figure 4) also highlighted the temporal complexity of the controls of height growth [32] implying the cumulative effects of meteorological conditions [21,56]. The model, though, showed low autocorrelation (Table 4), likely as the height increment was prewhitened [39,45], and the set of meteorological predictors indicated an extended window of responsiveness of height increment.
The dependence of growth responses on the baseline of weather conditions, i.e., the nonlinearity of response suitable for extrapolation beyond the calibration interval [37,38,57], was estimated for three of the eight regionally significant meteorological variables (Figure 5, Table 4). The prevalence of regional weather drivers with linear effects (Figure 5) implied the uniformity of weather sensitivity of height increment [9]. Still, the prevailing linear effects also suggested that the length of the studied climatic gradient has been limited with the respect to the range of conditions to which populations have adapted [26,27,37,38]. The prevailing linear responses might also be related to the spatial coverage of the datasets, which provided a higher representation of narrower climatic conditions in the south (Figure 1). Hence, these responses appear to be of limited extrapolation [9], although at a wider scale compared to those of radial increment [5].
Considering that studied stands grow under cold climates [58], the observed regional weather–growth relationships (Figure 5) indicated that the height increment of Scots pine is limited by low temperature (warmth), albeit that radial increment was sensitive to moisture availability [5,8,17,35,59]. On the other hand, warming has been mainly related to the dormancy period [40], when the temperature had a negative effect on height increment (Figure 5E), likely due to intensified respiration, reduction in nutrient reserves and premature dehardening [60,61]. The principal effects of meteorological conditions in the summer of the previous year (Table 4, Figure 5) comply with the dependence of height increment on growth initials, which form around the mid-summer [15,33].
In contrast to tree-ring width [4,5], temperature in the previous summer, particularly July, had a positive effect on height increment (Figure 5B,C), which can be explained by increased assimilation, and hence, the formation of additional growth initials and nutrient reserves [32,33,51,62]. This also implies that the thermal regime can modulate the allocation of assimilates to primary and secondary growth [16,19]. The nonlinear effect of temperature in the preceding June (Figure 5B) suggested that, already at the first part of the growing period, low temperature might affect (reduce) the formation of growth initials [32] via reduced assimilation and cell division [16,62,63]. The positive effect of October temperature, particularly in the warmest years/locations (Figure 5D), can be explained by the extension of the vegetation period and additional assimilation of nutrients [64], which would be deployed for the formation of the increment [17]. The positive effect of the minimum temperature in June–July of the year of increment (Figure 5F) highlighted the influence of thermal regime on the rate of cell division, and hence, the extension of growth initials [15,19,32]. The drought-related variables (Figure 4), however, showed local specifics, indicating stand-specifics in water availability, particularly under a milder climate of the Baltics [8,59].
The negative effects of meteorological conditions in the current August, which had the second highest effect on increment (Table 4, Figure 5G,H), are difficult to explain, as height growth is usually finalized at that time [16,65]. This might be related with the northernmost site (FIN_1), where growing season is timed later hence height growth might still finalize in August [16,62,65]. Perhaps, the negative effect of temperature might be related to heat stress and/or intrinsic water deficit, hence reduced increment [2,35]. The negative response to precipitation might be related to reduced temperature and solar radiation, and hence, assimilation [66], which, however, somewhat contradicts the negative effect of temperature. Alternatively, the effect of conditions in August might be related to Lammas growth, which is occasional and depends on the occurrence of favourable conditions [67,68], or the effect might be an artefact of the data. Nevertheless, the effects of temperature in August, which had a two-year lag support the cumulative influence of summer conditions on height increment [21,56].
The inter-annual variation of height increment and its estimated responses to weather conditions (Figure 4 and Figure 5), suggested the stationarity of climatic controls of growth, and hence, the reliability of their extrapolation [7,9], as local adaptation in terms of sensitivity was low. The estimated complex of weather–growth responses and age effects indicated that the anticipated climatic changes would likely increase the inter-annual variability of height increment [7,24]. Under such circumstances, the estimated plasticity of height increment, which, which is among the main determinants of sustainability [27,51,64], suggested the persistence of the adaptability of studied genotypes.

5. Conclusions

The study hypothesis was partially confirmed, as temperature was the main driving factor of height increment, while moisture availability generally showed weaker and local effects. The window of responsiveness of height increment indicated cumulative effects of meteorological conditions with environmental (climatic) signals being carried over up to two years. However, the observed regional weather–growth relationships were mostly linear and plastic (temporary variable), despite that bell-shaped responses were expected, thus suggesting that wide extrapolations might be still biased, particularly outside the climatic range of the north-eastern Baltic Sea region. Hence, the extension of the study to include data from sites growing under warmer climates, thus widening studied climatic gradient, would be necessary to estimate the adaptability of height growth to the anticipated environmental changes. Nevertheless, the prevailing linearity of the responses also suggested that climatic conditions have been suboptimal for height growth of Scots pine within the region, yet the analysed variability of those conditions was within the range to which the genotypes have adapted. The estimated complex regional controls over height growth, however, implied generally positive relationships with temperature, suggesting that increased increment in a warming climate could be contributing to the competitiveness and sustainability of Scots pine within the region.

Author Contributions

Conceptualization, R.M., S.M. and Ā.J.; methodology, R.M., S.M. and H.S.; software, R.K., A.K. and M.H.; validation, M.H., A.K. and H.S.; formal analysis, M.H. and S.M.; investigation, R.K., A.K., Ā.J., H.S. and S.M.; resources, Ā.J., A.K. and H.S.; data curation, R.K., M.H. and H.S.; writing—original draft preparation, R.M., S.M. and A.K.; writing—review and editing, H.S., S.M., M.H., R.M. and Ā.J.; visualization, R.M.; supervision, A.K. and Ā.J.; project administration, Ā.J., A.K. and H.S.; funding acquisition, Ā.J. All authors have read and agreed to the published version of the manuscript.

Funding

The study was funded by the European Regional Development Fund under the framework of the project “Decision support tool for increased forest productivity via efficient climate-adjusted transfer of genetic gain” (No. 1.1.1.1/19/A/111). The Estonian data collection was supported by the Estonia Estonian Ministry of Sciences (P180021, P180274, P200189; P200196), the Estonian Environmental Investment Centre, and the Estonian Research Council (PRG1007; PRG1674).

Institutional Review Board Statement

Not applicable.

Informed Consent Statement

Not applicable.

Data Availability Statement

Data available on request.

Acknowledgments

We acknowledge the technical staff, who helped to realize the study.

Conflicts of Interest

The authors declare no conflict of interest.

References

  1. Buras, A.; Menzel, A. Projecting tree species composition changes of European forests for 2061-2090 under RCP 4.5 and RCP 8.5 Scenarios. Front. Plant Sci. 2019, 9, N1986. [Google Scholar] [CrossRef] [PubMed] [Green Version]
  2. Isaac-Renton, M.; Montwe, D.; Hamann, A.; Spiecker, H.; Cherubini, P.; Treydte, K. Northern forest tree populations are physiologically maladapted to drought. Nat. Commun. 2018, 9, 5254. [Google Scholar] [CrossRef] [PubMed] [Green Version]
  3. Aitken, S.N.; Bemmels, J.B. Time to get moving: Assisted gene flow of forest trees. Evol. Appl. 2016, 9, 271–290. [Google Scholar] [CrossRef] [PubMed]
  4. Matisons, R.; Elferts, D.; Krišāns, O.; Schneck, V.; Gärtner, H.; Bast, A.; Wojda, T.; Kowalczyk, J.; Jansons, Ā. Non-linear regional weather-growth relationships indicate limited adaptability of the eastern Baltic Scots pine. For. Ecol. Manag. 2021, 479, N118600. [Google Scholar] [CrossRef]
  5. Matisons, R.; Elferts, D.; Krišāns, O.; Schneck, V.; Gärtner, H.; Wojda, T.; Kowalczyk, J.; Jansons, Ā. Nonlinear weather-growth relationships suggest disproportional growth changes of Norway spruce in the eastern Baltic region. Forests 2021, 12, 661. [Google Scholar] [CrossRef]
  6. Cavin, L.; Jump, A.S. Highest drought sensitivity and lowest resistance to growth suppression are found in the range core of the tree Fagus sylvatica L. not the equatorial range edge. Glob. Chang. Biol. 2017, 23, 362–379. [Google Scholar] [CrossRef] [Green Version]
  7. Huber, N.; Bugmann, H.; Cailleret, M.; Bircher, N.; Lafond, V. Stand-scale climate change impacts on forests over large areas: Transient responses and projection uncertainties. Ecol. Appl. 2021, 31, e02313. [Google Scholar] [CrossRef]
  8. Harvey, J.; Smiljanić, M.; Scharnweber, T.; Buras, A.; Cedro, A.; Cruz-García, R.; Drobyshev, I.; Janecka, K.; Jansons, Ā.; Kaczka, R.; et al. Tree growth influenced by warming winter climate and summer moisture availability in northern temperate forests. Glob. Chang. Biol. 2020, 26, 2505–2518. [Google Scholar] [CrossRef]
  9. Wilmking, M.; van der Maaten-Theunissen, M.; van der Maaten, E.; Scharnweber, T.; Buras, A.; Biermann, C.; Gurskaya, M.; Hallinger, M.; Lange, J.; Shetti, R.; et al. Global assessment of relationships between climate and tree growth. Glob. Chang. Biol. 2020, 26, 3212–3220. [Google Scholar] [CrossRef] [PubMed] [Green Version]
  10. Kalliokoski, T.; Mäkelä, A.; Fronzek, S.; Minunno, F.; Peltoniemi, M. Decomposing sources of uncertainty in climate change projections of boreal forest primary production. Agron. For. Meteorol. 2018, 262, 192–205. [Google Scholar] [CrossRef]
  11. Zhang, Z.; Babst, F.; Bellassen, V.; Frank, D.; Launois, T.; Tan, K.; Ciais, P.; Poulter, B. Converging climate sensitivities of European forests between observed radial tree growth and vegetation models. Ecosystems 2018, 21, 410–425. [Google Scholar] [CrossRef] [Green Version]
  12. Xu, K.; Wang, X.; Liang, P.; An, H.; Sun, H.; Han, W.; Li, Q. Tree-ring widths are good proxies of annual variation in forest productivity in temperate forests. Sci. Rep. 2017, 7, 1945. [Google Scholar] [CrossRef] [PubMed] [Green Version]
  13. Breed, M.F.; Harrison, P.A.; Bischoff, A.; Durruty, P.; Gellie, N.J.C.; Gonzales, E.K.; Havens, K.; Karmann, M.; Kilkenny, F.F.; Krauss, S.L.; et al. Priority actions to improve provenance decision-making. BioScience 2018, 68, 510–516. [Google Scholar] [CrossRef]
  14. Housset, J.M.; Nadeau, S.; Isabel, N.; Depardieu, C.; Duchesne, I.; Lenz, P.; Girardin, M.P. Tree rings provide a new class of phenotypes for genetic associations that foster insights into adaptation of conifers to climate change. New Phytol. 2018, 218, 630–645. [Google Scholar] [CrossRef] [Green Version]
  15. Salminen, H.; Jalkanen, R. Modelling the effect of temperature on height increment of Scots pine at high latitudes. Silva Fenn. 2005, 39, 497–508. [Google Scholar] [CrossRef] [Green Version]
  16. Mäkinen, H.; Jyske, T.; Nöjd, P. Dynamics of diameter and height increment of Norway spruce and Scots pine in southern Finland. Ann. For. Sci. 2018, 75, N28. [Google Scholar] [CrossRef] [Green Version]
  17. Jansons, Ā.; Matisons, R.; Zadiņa, M.; Sisenis, L.; Jansons, J. The effect of climatic factors on height increment of Scots pine in sites differing by continentality in Latvia. Silva Fenn. 2015, 49, N1262. [Google Scholar] [CrossRef] [Green Version]
  18. Salminen, H.; Jalkanen, R. Intra-annual height increment of Pinus sylvestris at high latitudes in Finland. Tree Physiol. 2007, 27, 1347–1353. [Google Scholar] [CrossRef] [Green Version]
  19. Kellomäki, S.; Ikonen, V.P.; Peltola, H.; Kolström, T. Modelling the structural growth of Scots pine with implications for wood quality. Ecol. Model. 1999, 122, 117–134. [Google Scholar] [CrossRef]
  20. Loehle, C. Height growth rate tradeoffs determine northern and southern range limits for trees. J. Biogeogr. 1998, 25, 735–742. [Google Scholar] [CrossRef]
  21. Metslaid, S.; Sims, A.; Kangur, A.; Hordo, M.; Jogiste, K.; Kiviste, A.; Hari, P. Growth patterns from different forest generations of Scots pine in Estonia. J. For. Res. 2011, 16, 237–243. [Google Scholar] [CrossRef]
  22. Mäkinen, H. The suitability of height and radial increment variation in Pinus sylvestris (L.) for expressing environmental signals. For. Ecol. Manag. 1998, 112, 191–197. [Google Scholar] [CrossRef]
  23. Nunes, L.; Patricio, M.; Tome, J.; Tome, M. Modeling dominant height growth of maritime pine in Portugal using GADA methodology with parameters depending on soil and climate variables. Ann. For. Sci. 2011, 68, 311–323. [Google Scholar] [CrossRef] [Green Version]
  24. Pensa, M.; Salminen, H.; Jalkanen, R. A 250-year-long height-increment chronology for Pinus sylvestris at the northern coniferous timberline: A novel tool for reconstructing past summer temperatures? Dendrochronologia 2005, 22, 75–81. [Google Scholar] [CrossRef]
  25. Reich, P.B.; Oleksyn, J. Climate warming will reduce growth and survival of Scots pine except in the far north. Ecol. Lett. 2008, 11, 588–597. [Google Scholar] [CrossRef]
  26. Nabais, C.; Hansen, J.K.; David-Schwartz, R.; Klisz, M.; López, R.; Rozenberg, P. The effect of climate on wood density: What provenance trials tell us? For. Ecol. Manag. 2018, 408, 148–156. [Google Scholar] [CrossRef]
  27. Moran, E.; Lauder, J.; Musser, C.; Stathos, A.; Shu, M. The genetics of drought tolerance in conifers. New Phytol. 2017, 216, 1034–1048. [Google Scholar] [CrossRef] [Green Version]
  28. Way, D.A.; Oren, R. Differential responses to changes in growth temperature between trees from different functional groups and biomes: A review and synthesis of data. Tree Physiol. 2010, 30, 669–688. [Google Scholar] [CrossRef] [Green Version]
  29. Nabuurs, G.J.; Verkerk, P.J.; Schelhaas, M.J.; González Olabarria, J.R.; Trasobares, A.; Cienciala, E. Climate-Smart Forestry: Mitigation Impacts in Three European Regions. European Forest Institute. 2018. Available online: https://www.efi.int/sites/default/files/files/publication-bank/2018/efi_fstp_6_2018.pdf (accessed on 15 February 2020).
  30. Jansons, Ā.; Matisons, R.; Baumanis, I.; Puriņa, L. Effect of climatic factors on height increment of Scots pine in experimental plantation in Kalsnava, Latvia. For. Ecol. Manag. 2013, 306, 185–191. [Google Scholar] [CrossRef]
  31. Oleksyn, J.; Reich, P.B.; Tjoelker, M.G.; Chalupka, W. Biogeographic differences in shoot elongation pattern among European Scots pine populations. For. Ecol. Manag. 2001, 148, 207–220. [Google Scholar] [CrossRef]
  32. Lanner, R.M. Patterns of shoot development in Pinus and their relationship to growth potential. In Tree Physiology and Yield Improvement; Cannell, M.G.R., Last, F.T., Eds.; Academic Press: London, UK, 1976; pp. 223–243. [Google Scholar]
  33. Lanner, R.M. On the insensitivity of height growth to spacing. For. Ecol. Manag. 1985, 13, 143–148. [Google Scholar] [CrossRef]
  34. Friedrichs, D.A.; Büntgen, U.; Frank, D.C.; Esper, J.; Neuwirth, B.; Löffler, J. Complex climate controls on 20th century oak growth in Central-West Germany. Tree Physiol. 2009, 29, 39–51. [Google Scholar] [CrossRef] [PubMed] [Green Version]
  35. Allen, C.D.; Breshears, D.D.; McDowell, N.G. On underestimation of global vulnerability to tree mortality and forest die-off from hotter drought in the Anthropocene. Ecosphere 2015, 6, 1–55. [Google Scholar] [CrossRef]
  36. Choat, B.; Jansen, S.; Brodribb, T.J.; Cochard, H.; Delzon, S.; Bhaskar, R.; Bucci, S.J.; Feild, T.S.; Gleason, S.M.; Hacke, U.G.; et al. Global convergence in the vulnerability of forests to drought. Nature 2012, 491, 752–755. [Google Scholar] [CrossRef] [Green Version]
  37. Lloyd, A.H.; Duffy, P.A.; Mann, D.H. Nonlinear responses of white spruce growth to climate variability in interior Alaska. Can. J. For. Res. 2013, 43, 331–343. [Google Scholar] [CrossRef]
  38. Lloyd, A.H.; Bunn, A.G.; Berner, L. A latitudinal gradient in tree growth response to climate warming in the Siberian taiga. Glob. Chang. Biol. 2011, 17, 1935–1945. [Google Scholar] [CrossRef]
  39. Fei, S.; Desprez, J.M.; Potter, K.M.; Jo, I.; Knott, J.A.; Oswalt, C.M. Divergence of species responses to climate change. Sci. Adv. 2017, 3, e1603055. [Google Scholar] [CrossRef] [Green Version]
  40. Meier, H.E.M.; Kniebusch, M.; Dieterich, C.; Gröger, M.; Zorita, E.; Elmgren, R.; Myrberg, K.; Ahola, M.P.; Bartosova, A.; Bonsdorff, E.; et al. Climate change in the Baltic Sea region: A summary. Earth Syst. Dynam. Discuss. 2022, 13, 457–593. [Google Scholar] [CrossRef]
  41. Harris, I.; Osborn, T.J.; Jones, P.; Lister, D. Version 4 of the CRU TS monthly high-resolution gridded multivariate climate dataset. Sci. Data 2020, 7, 109. [Google Scholar] [CrossRef] [Green Version]
  42. Vicente-Serrano, S.M.; Begueria, S.; Lopez-Moreno, J.I. A multiscalar drought index sensitive to global warming: The standardized precipitation evapotranspiration index—SPEI. J. Clim. 2010, 23, 1696–1718. [Google Scholar] [CrossRef] [Green Version]
  43. Lanner, R.M. Development of the terminal bud and shoot of slash pine saplings. For. Sci. 1978, 24, 167–179. [Google Scholar]
  44. Holmes, R. Computer-assisted quality control in tree-ring dating and measurement. Tree-Ring Bull. 1983, 43, 69–78. [Google Scholar]
  45. Cook, E.R. The decomposition of tree-ring series for environmental studies. Tree-Ring Bull. 1987, 47, 37–59. [Google Scholar]
  46. Bunn, A.G. A dendrochronology program library in R (dplR). Dendrochronologia 2008, 26, 115–124. [Google Scholar] [CrossRef]
  47. Wigley, T.M.L.; Briffa, K.R.; Jones, P.D. On the average value of correlated time series, with applications in dendroclimatology and hydrometeorology. J. Clim. Appl. Meteorol. 1984, 23, 201–213. [Google Scholar] [CrossRef]
  48. Zang, C.; Biondi, F. treeclim: An R package for the numerical calibration of proxy-climate relationships. Ecography 2015, 38, 431–436. [Google Scholar] [CrossRef]
  49. Wood, S.N. Fast stable restricted maximum likelihood and marginal likelihood estimation of semiparametric generalized linear models. J. R. Stat. Soc. B 2011, 73, 3–36. [Google Scholar] [CrossRef] [Green Version]
  50. R Core Team. R: A Language and Environment for Statistical Computing; R Foundation for Statistical Computing: Vienna, Austria. 2022. Available online: https://www.R-project.org/ (accessed on 26 October 2022).
  51. von Arx, G.; Arzac, A.; Fonti, P.; Frank, D.; Zweifel, R.; Rigling, A.; Galiano, L.; Gessler, A.; Olano, J.M. Responses of sapwood ray parenchyma and non-structural carbohydrates of Pinus sylvestris to drought and long-term irrigation. Funct. Ecol. 2017, 31, 1371–1382. [Google Scholar] [CrossRef] [Green Version]
  52. Restaino, C.M.; Peterson, D.L.; Littell, J. Increased water deficit decreases Douglas fir growth throughout western US forests. Proc. Nat. Acad. Sci. USA 2016, 113, 9557–9562. [Google Scholar] [CrossRef] [Green Version]
  53. Prendin, A.L.; Mayr, S.; Beikircher, B.; von Arx, G.; Petit, G. Xylem anatomical adjustments prioritize hydraulic efficiency over safety as Norway spruce trees grow taller. Tree Physiol. 2018, 38, 1088–1097. [Google Scholar] [CrossRef]
  54. Buras, A.; Schunk, C.; Zeiträg, C.; Herrmann, C.; Kaiser, L.; Lemme, H.; Straub, C.; Taeger, S.; Gößwein, S.; Klemmt, H.-J.; et al. Are Scots pine forest edges particularly prone to drought-induced mortality? Environ. Res. Lett. 2018, 13, N025001. [Google Scholar] [CrossRef]
  55. Hacket-Pain, A.; Ascoli, D.; Berretti, R.; Mencuccini, M.; Motta, R.; Nola, P.; Piussi, P.; Ruffinatto, F.; Vacchiano, G. Temperature and masting control Norway spruce growth, but with high individual tree variability. For. Ecol. Manag. 2019, 438, 142–150. [Google Scholar] [CrossRef] [Green Version]
  56. Lloret, F.; Keeling, E.G.; Sala, A. Components of tree resilience: Effects of successive low-growth episodes in old ponderosa pine forests. Oikos 2011, 120, 1909–1920. [Google Scholar] [CrossRef]
  57. Billings, S.A.; Glaser, S.M.; Boone, A.S.; Stephen, F.M. Nonlinear tree growth dynamics predict resilience to disturbance. Ecosphere 2015, 6, 1–13. [Google Scholar] [CrossRef] [Green Version]
  58. Kottek, M.; Grieser, J.; Beck, C.; Rudolf, B.; Rubel, F. World map of the Köppen-Geiger climate classification updated. Meteorol. Z. 2006, 15, 259–263. [Google Scholar] [CrossRef]
  59. Henttonen, H.M.; Mäkinen, H.; Heiskanen, J.; Peltoniemi, M.; Lauren, A.; Hordo, M. Response of radial increment variation of Scots pine to temperature, precipitation and soil water content along a latitudinal gradient across Finland and Estonia. Agric. For. Meteorol. 2014, 198, 294–308. [Google Scholar] [CrossRef]
  60. Beck, E.H.; Heim, R.; Hansen, J. Plant resistance to cold stress: Mechanisms and environmental signals triggering frost hardening and dehardening. J. Biosci. 2004, 29, 449–459. [Google Scholar] [CrossRef]
  61. Ögren, E. Relationship between temperature, respiratory loss of sugar and premature dehardening in dormant Scots pine seedlings. Tree Physiol. 1997, 17, 47–51. [Google Scholar] [CrossRef]
  62. Jyske, T.; Mäkinen, H.; Kalliokoski, T.; Nöjd, P. Intra-annual tracheid production of Norway spruce and Scots pine across a latitudinal gradient in Finland. Agron. For. Meteorol. 2014, 194, 241–254. [Google Scholar] [CrossRef]
  63. Cuny, H.E.; Fonti, P.; Rathgeber, C.B.; von Arx, G.; Peters, R.L.; Frank, D.C. Couplings in cell differentiation kinetics mitigate air temperature influence on conifer wood anatomy. Plant Cell Environ. 2019, 4, 1222–1232. [Google Scholar] [CrossRef]
  64. White, M.A.; Running, S.W.; Thornton, P.E. The impact of growing-season length variability on carbon assimilation and evapotranspiration over 88 years in the eastern US deciduous forest. Int. J. Biometeorol. 1999, 42, 139–145. [Google Scholar] [CrossRef] [PubMed]
  65. Pensa, M.; Sepp, M.; Jalkanen, R. Connections between climatic variables and the growth and needle dynamics of Scots pine (Pinus sylvestris L.) in Estonia and Lapland. Int. J. Biometeorol. 2006, 50, 205–214. [Google Scholar] [CrossRef] [PubMed]
  66. Strand, M.; Löfvenius, M.O.; Bergsten, U.; Lundmark, T.; Rosvall, O. Height growth of planted conifer seedlings in relation to solar radiation and position in Scots pine shelterwood. For. Ecol. Manag. 2006, 224, 258–265. [Google Scholar] [CrossRef]
  67. Katrevics, J.; Neimane, U.; Dzerina, B.; Kitenberga, M.; Jansons, J.; Jansons, A. Environmental factors affecting formation of lammas shoots in young stands of Norway spruce (Picea abies Karst.) in Latvia. iForest 2018, 11, 809–815. [Google Scholar] [CrossRef] [Green Version]
  68. Jansons, A.; Neimane, U.; Dzerina, B.; Adamovics, A. Influence of lammas shoots on height of young Scots pines in Latvia. Agron. Res. 2016, 14, 407–417. [Google Scholar]
Forests 14 00791 g001 550
Figure 1. Location of the studied stands of Scots pine represented by the time series of height increment in the Baltic Sea region.
Figure 1. Location of the studied stands of Scots pine represented by the time series of height increment in the Baltic Sea region.
Forests 14 00791 g001
Forests 14 00791 g002 550
Figure 2. Cross-dated time series of height increment of Scots pine from conventionally managed stands in Finland, Estonia and Latvia and the mean series (bold line) of the datasets of the studied stands.
Figure 2. Cross-dated time series of height increment of Scots pine from conventionally managed stands in Finland, Estonia and Latvia and the mean series (bold line) of the datasets of the studied stands.
Forests 14 00791 g002
Forests 14 00791 g003 550
Figure 3. Residual chronologies (black line) of height increment of Scots pine from conventionally managed stands in Finland, Estonia and Latvia. Grey line indicates replication of annual data represented by the chronologies.
Figure 3. Residual chronologies (black line) of height increment of Scots pine from conventionally managed stands in Finland, Estonia and Latvia. Grey line indicates replication of annual data represented by the chronologies.
Forests 14 00791 g003
Forests 14 00791 g004 550
Figure 4. Bootstrapped Pearson correlation coefficients between residual chronologies of height increment of Scots pine from conventionally managed stands in Northern Finland, Estonia and Latvia and meteorological variables (monthly standardized precipitation evapotranspiration indices SPEI, precipitation sums, and mean temperature). bepr. and prev. denote meteorological conditions two and one year before the formation of increment, respectively. The significant correlations (at α = 0.05) are shown in colour. The analysis was conducted for the common period of 1955–2003.
Figure 4. Bootstrapped Pearson correlation coefficients between residual chronologies of height increment of Scots pine from conventionally managed stands in Northern Finland, Estonia and Latvia and meteorological variables (monthly standardized precipitation evapotranspiration indices SPEI, precipitation sums, and mean temperature). bepr. and prev. denote meteorological conditions two and one year before the formation of increment, respectively. The significant correlations (at α = 0.05) are shown in colour. The analysis was conducted for the common period of 1955–2003.
Forests 14 00791 g004
Forests 14 00791 g005 550
Figure 5. The estimated effects of meteorological variables (A–H) on the relative additional height increment of Scots pine (as represented by the residual chronologies) of conventionally managed stands across the north-eastern Baltic Sea region in Finland, Estonia, and Latvia. The dotted lines/shaded regions show 95% confidence intervals of the responses. Panels are ordered chronologically. bepr.—year before previous, prev.—previous year.
Figure 5. The estimated effects of meteorological variables (A–H) on the relative additional height increment of Scots pine (as represented by the residual chronologies) of conventionally managed stands across the north-eastern Baltic Sea region in Finland, Estonia, and Latvia. The dotted lines/shaded regions show 95% confidence intervals of the responses. Panels are ordered chronologically. bepr.—year before previous, prev.—previous year.
Forests 14 00791 g005
Table
Table 1. Location and description of the studied conventionally managed stands and dimension of sampled tress used for the generalization of regional responses of height increment of Scots pine to meteorological conditions across the north-eastern Baltic Sea region. The refences of initial publication of the datasets are given. DBH—stem diameter at breast height, H—tree height.
Table 1. Location and description of the studied conventionally managed stands and dimension of sampled tress used for the generalization of regional responses of height increment of Scots pine to meteorological conditions across the north-eastern Baltic Sea region. The refences of initial publication of the datasets are given. DBH—stem diameter at breast height, H—tree height.
FIN_1EST_1EST_2EST_3LAT_1LAT_2
LocalityNorthern Finland, RovaniemiEastern part of Estonia, JarvseljaEastern part of Estonia, JarvseljaEastern part of Estonia, JarvseljaWestern part of Latvia, SkedeEastern part of Latvia, Kalsnava
Lat., °N66.3858.3258.3258.3157.2656.58
Lon., °E26.6727.2427.2527.3022.6925.94
SoilMesotrophic, dryish heathEutrophic, mineral free drainingEutrophic, mineral free drainingEutrophic, mineral free drainingMesotrophic, mineral free drainingMesotrophic, mineral free draining
Stand originNaturally regeneratedSeededSeededSeededNaturally regeneratedNaturally regenerated
Age29–35113–11653–7539–43104–10878–87
DBH (±st. dev.), cm13.8 ± 0.844.0 ± 5.731.6 ± 3.229.7 ± 5.342.3 ± 6.139.8 ± 5.9
H (±st. dev.), m9.8 ± 0.732.3 ± 2.428.8 ± 1.024.0 ± 0.829.9 ± 2.527.4 ± 1.7
Trees sampled201013102020
Reference[15][21][21][21][17][17]
Table
Table 2. Climatic description of the studied conventionally managed stands of Scots pine according to the gridded climatic data. Temperature is in °C and precipitation is in mm.
Table 2. Climatic description of the studied conventionally managed stands of Scots pine according to the gridded climatic data. Temperature is in °C and precipitation is in mm.
FIN_1EST_1-3LAT_1LAT_2
Mean annual temperature (±st. deviation)0.8 ± 1.15.4 ± 1.06.3 ± 0.95.8 ± 0.8
Mean minimum February temperature −15.9−9.1−5.8−7.9
Mean maximum February temperature−6.8−2.5−0.5−1.7
Mean February temperature−11.3−5.8−3.1−4.8
Mean minimum July temperature 10.311.912.111.8
Mean maximum July temperature20.722.520.722.1
Mean July temperature15.517.216.416.9
Mean minimum May-September temperature6.28.99.59.1
Mean maximum May-September temperature15.919.318.019.3
Mean May-September temperature11.014.113.714.2
Mean annual precipitation sum (±st. deviation)495 ± 71618 ± 80691 ± 93673 ± 80
Mean annual precipitation sum (±st. deviation)271 ± 53341 ± 80322 ± 70346 ± 72
Table
Table 3. Statistics of the cross-dated datasets of time series of height increment (HI) of Scots pine from the stands from the north-eastern Baltic Sea region.
Table 3. Statistics of the cross-dated datasets of time series of height increment (HI) of Scots pine from the stands from the north-eastern Baltic Sea region.
FIN_1EST_1EST_2EST_3LAT_1LAT_2
Timespan1953–20031891–20081952–20081966–20091906–20121927–2013
Mean HI, cm32.328.550.158.033.634.3
St. dev. HI, cm8.218.616.715.215.615.3
Number of time series201310101513
Mean interseries correlation0.470.280.560.580.340.29
Expressed population signal0.880.860.890.870.880.85
Signal-to-noise ratio7.426.048.036.487.064.64
First order autocorrelation0.320.400.380.430.340.21
Gini coefficient0.130.170.110.100.130.12
Mean sensitivity 0.210.270.170.150.220.2
Mean synchronicity (Gleichläufigkeit)0.700.640.710.720.610.62
Table
Table 4. Statistics of the regional generalization of responses of height increment of Scots pine from conventionally managed stands to meteorological conditions. For the fixed effects (meteorological variables), the effective degrees of freedom indicating the shape of the responses, their strength (F-value) and significance (p-value) are shown; for the random effects, estimated variance is shown. The overall performance of the model is described by marginal and conditional R2, as well as root mean square errors (REML) during calibration and cross verification. Significance codes; p-values: * < 0.05, ** < 0.01, *** < 0.001. Prev.—previous year, bepr.—before previous year.
Table 4. Statistics of the regional generalization of responses of height increment of Scots pine from conventionally managed stands to meteorological conditions. For the fixed effects (meteorological variables), the effective degrees of freedom indicating the shape of the responses, their strength (F-value) and significance (p-value) are shown; for the random effects, estimated variance is shown. The overall performance of the model is described by marginal and conditional R2, as well as root mean square errors (REML) during calibration and cross verification. Significance codes; p-values: * < 0.05, ** < 0.01, *** < 0.001. Prev.—previous year, bepr.—before previous year.
Fixed Effects (Smoothing Terms)Effective Degree of FreedomF-Value
Mean temperature. bepr. August1.398.87 ***
Mean temperature prev. June2.137.11 **
Mean temperature prev. July1.0015.21 ***
Mean temperature prev. October2.0610.39 ***
Mean temperature January–February1.005.09 *
Minimum mean June–July temperature1.004.66 *
Mean temperature August1.0011.90 ***
Precipitation sum August1.006.83 **
Random Effects (Variance)
Stand (site)3.49 × 10−5
Year1.23 × 10−2
Residual8.91 × 10−7
Model Performance
Autocorrelation term0.02
Marginal R20.38
Conditional R20.72
REML (calibration)0.11
REML (verification)0.14
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

Share and Cite

MDPI and ACS Style

Matisons, R.; Metslaid, S.; Hordo, M.; Kask, R.; Kangur, A.; Salminen, H.; Jansons, Ā. Direct and Carry-Over Effects of Temperature Drive Height Increment of Scots Pine in the North-Eastern Baltic Sea Region. Forests 2023, 14, 791. https://doi.org/10.3390/f14040791

AMA Style

Matisons R, Metslaid S, Hordo M, Kask R, Kangur A, Salminen H, Jansons Ā. Direct and Carry-Over Effects of Temperature Drive Height Increment of Scots Pine in the North-Eastern Baltic Sea Region. Forests. 2023; 14(4):791. https://doi.org/10.3390/f14040791

Chicago/Turabian Style

Matisons, Roberts, Sandra Metslaid, Maris Hordo, Regino Kask, Ahto Kangur, Hannu Salminen, and Āris Jansons. 2023. "Direct and Carry-Over Effects of Temperature Drive Height Increment of Scots Pine in the North-Eastern Baltic Sea Region" Forests 14, no. 4: 791. https://doi.org/10.3390/f14040791

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Metrics

Back to TopTop