Norway spruce responses to drought forcing in areas affected by forest decline

Petr Čermák (Čermák, P)

Faculty of Forestry and Wood Technology, Mendel University in Brno, Zemědělská 3, 613 00 Brno, Czech Republic;

Tomáš Kolář (Kolář, T)

Faculty of Forestry and Wood Technology, Mendel University in Brno, Zemědělská 3, 613 00 Brno, Czech Republic;

Global Change Research Institute of the Czech Academy of Sciences, Bělidla 986/4a, 603 00 Brno, Czech Republic

Tomáš Žid (Žid, T)

Faculty of Forestry and Wood Technology, Mendel University in Brno, Zemědělská 3, 613 00 Brno, Czech Republic;

Miroslav Trnka (Trnka, M)

Global Change Research Institute of the Czech Academy of Sciences, Bělidla 986/4a, 603 00 Brno, Czech Republic

Michal Rybníček (Rybníček, M)

Faculty of Forestry and Wood Technology, Mendel University in Brno, Zemědělská 3, 613 00 Brno, Czech Republic;

Global Change Research Institute of the Czech Academy of Sciences, Bělidla 986/4a, 603 00 Brno, Czech Republic



Aim of study: To assess the crown condition and radial growth of Norway spruce in plots with an increasing frequency and strength of drought during the last decades.

Area of study: Northern Moravia, Czech Republic.

Materials and methods: Crown condition assessment and dendrochronology analysis were used.

Main results: Tree-ring width (TRW) was significantly influenced by previous autumn and current summer climate. The temporal variability of the growth-climate relationship shows that the impact of water sufficiency (precipitation, relative soil water content, drought index) markedly increased mainly during the 2000s and the 2010s. Most climate-growth relationships were significant only in the last two or three decades. The observed crown conditions and their relationships with TRW also indicate stress intensification during the same period. Our results suggest that water availability was the main factor affecting radial growth and the occurrence of negative pointer years and was probably also the factor triggering the decline.

Research highlights: In these current site and climate conditions, the silviculture of Norway spruce is extremely risky in the study area. Our results have also shown that the observed climate change is too dynamic for long-term forest plans, especially with regard to recommended forest species composition.

Additional Keywords: defoliation; Picea abies; tree-ring width; precipitation; PDSI; available soil water.

Authors' contributions: Conceived and designed the experiments: PČ, TK, MR. Performed the experiments: PČ, TK, TŽ and MR. Analysed the data: PČ, TK, MT, MR. Contributed reagents/materials/analysis tools: MT. Wrote the paper: PČ and TK.

Citation: Čermák, P., Kolář, T., Žid, T., Trnka, M., Rybníček, M. (2019). Norway spruce responses to drought forcing in areas affected by forest decline. Forest Systems, Volume 28, Issue 3, e016.

Received: 13 Mar 2019. Accepted: 31 Oct 2019.

Copyright © 2019 INIA. This is an open access article distributed under the terms of the Creative Commons Attribution 4.0 International (CC-by 4.0) License.

Funding agencies/Institutions Project/Grant
EEA Grants project “Frameworks and possibilities of forest adaptation measures and strategies connected with climate change” EHP-CZ02-OV-1-019-2014
Ministry of Education, Youth and Sports of CR within the National Sustainability Program I (NPU I) LO1415
Czech Republic Grant Agency 18-17295S

Competing interests: The authors have declared that no competing interests exist.

Correspondence should be addressed to Petr Čermák: e-mail





Material and Methods






Tree mortality usually involves complex processes and multiple interactions among disturbance agents and stress factors occurring at different time scales (Franklin et al., 1987; Trumbore et al., 2015). Tree death can be caused by abiotic stress factors interacting with different biotic factors, resulting in forest decline (Manion, 1991; Manion & Lachance, 1992; Sturrock et al. 2011). Drought is an important and frequent causal factor of forest decline; it can function as a predisposing, inciting or contributing factor (Sinclair, 1967). Forest resilience to drought mainly depends on drought intensity, but it may also depend on tree species or specific functional traits (Greenwood et al., 2017); furthermore, it can vary from one individual to another (Gazol et al., 2017; Camarero et al., 2018). Picea abies is one of the most drought-sensitive European tree species (van der Maaten-Theunissen et al., 2013; Hartl-Meier et al., 2014; Leuschner & Ellenberg, 2017) because it is a water-demanding species often having a sha-llow root system. During recent decades, drought-triggered forest declines have appeared in different forest types and climate zones. Allen et al. (2010) identified 88 episodes of drought and/or heat-induced forest mortality over the past years, and 10 of them affected the Picea spp.

One of the last Picea abies forest decline episodes started at the beginning of the 1990s in the Po-lish part of the Silesian Beskids Mts., and after the dry year 2003, it has spread southward to the Czech Republic (Rybníček et al., 2010). The main identified predisposing factors of forest decline were the artificial origin of stands, low content of basic elements in the soil, and mainly drought spells, especially in 2003 and 2007 (Novotný et al., 2008; Šrámek et al., 2008; Rybníček et al., 2010; Cienciala et al., 2017). Severe infestation by fungi Armillaria spp. was often an inciting factor for a subsequent attack of bark beetles (Ips typographus, Ips duplicatus and Pityogenes chalcographus) as a contributing factor causing final mortality (Grodzki, 2010; Vakula et al., 2015). Moreover, root decays caused by Armillaria spp. and some other primary pa-rasitic wood-destroying fungi increase the host water deficit and predispose it to the invasion of bark beetles (Jankovský et al., 2003; Grodzki, 2007; Holuša et al., 2018).

The stress response of trees in radial growth and crown conditions may occur with varying delays. In some cases, the first observable reaction to a stressor is discoloration or defoliation, and in other cases, the growth reduction is immediate, whereas the change in crown can be observed later. The differences in reaction may be caused by stress intensity, its du-ration or kind of affecting stressor (Rybníček et al., 2012a). Therefore, the use of both indicators is suitable for a precise description of forest decline processes.

With respect to the recurrent serious drought spells in the last two decades, we assessed the crown condition and radial growth of surviving Picea abies trees in northern Moravia (Oderské vrchy or Odra Highlands), an area of Central Europe that is strongly affected by forest decline. Our initial hypothesis was that the changing climate conditions, especially the rapid decrease in soil water availability, considerably affected crown conditions and tree ring width (TRW). Our aims were to i) investigate possible relationships between crown condition parameters and tree-ring width and ii) analyse changes in climate-growth re-lationships during a period of progressive climate change. 

Material and MethodsTop

Study plots

The study area is a plateau in the submontane belt in the Oderské vrchy, or Odra Highlands. We selected six circular plots (min. r = 32 m) in a forest district with the presence of declining stands. The plots were selected randomly in the prevailing site conditions (i.e., prevailing altitudinal zones and edaphic categories in the study area). More detailed information about the plots is provided in Table 1. All study plots were either Norway spruce artificial monocultures or stands with spruce as a dominant tree species (> 90% in species composition). The assessed stands have a closed canopy. The same silvicultural treatments were used in all plots during the stands’ history, repeating thinning from below at ages up to 40 years and eliminating unhealthy trees from older stands. The plots are representative of Norway spruce stands of the study area.

Table 1. Parameters of study plots.

The dominant soils were spodo-dystric cambisols. The soil condition of the area was evaluated by Šrámek et al. (2015). Soil profiles were acidified with a median pH (CaCl2) < 4 in the organic layer and the upper 30 cm of mineral soil. The strong deficiency of exchangeable calcium (< 140 has been found to the depth of 50 cm and magnesium deficiency (< 40 to the depth of 60 cm. The total nitrogen content is sufficient or good within the soil profile.

Climate conditions

Station climate data covering the period 1961–2014 were derived through interpolation from a set of nearby weather stations using locally weighted regressions, including the effect of altitude. The in-terpolation was based on the database of the Czech Hydrometeorological Institute and included me-asurements from 268 meteorological stations and 787 precipitation stations representing the territory of the Czech Republic.

We tested all observed weather variables for out-liers and breaks through a detailed homogenization sequence, and missing data were filled using the methods described by Štěpánek et al., (2009 and 2011). We then employed the SoilClim (Hlavinka et al., 2011) model to estimate the daily values of the relative soil water content (AWR) for the top 1.3 m, which was used as one of the water availability proxies, with 0 representing the wilting point and 100 the field capacity. The SoilClim model divided the soil profile into 2 layers (0–40 cm and 40 to 130 cm), which were also considered TRW predictors but were found to perform worse than the relative available water content in the entire profile. We also calculated the Palmer drought severity index (PDSI) using the procedure described by Palmer (1965).

The climate conditions changed in the study plots during the observed period (Fig. 1). Average temperatures increased, precipitation totals rapidly fluctuated annually without significant changes in mid-term average totals (Table 2), and AWR decreased, especially the summer AWR (Fig. 1). Severe drought in the growing season (April–September) repeated approximately once every ten years, but the last two dry episodes (2003–2007 and 2012–2014) were multiannual, in contrast to the previous droughts (Fig. 1). The average interval between two drought years was markedly shorter, and the average length of the drought period was longer in the 1991–2014 period compared to the 1961–1990 period (Fig. 2).

Figure 1. Climate development in the growing season during 1961–2014. The grey lines represent the March–May period, and the black lines represent the June–August period. The dotted lines are polynomial trendlines.

Table 2. Average values of temperature and precipitation during the growing season and the entire year.

Figure 2. Mean interval between drought seasons (in years) and number of drought seasons per 10 years calculated separately March–May (Mar–May), June–August (Jun–Aug), September– November (Sep–Nov) and December–February (Dec–Feb) and both parts (1961–1990 and 1991–2014) of the study period. A season was considered a drought season when the PDSI values were below -2 during the whole season.

Crown condition assessment

The plot design is the same as that of Level I of the ICP Forest network – a circular plot with a fixed number of trees (Eichhorn et al. 2010; Ferretti & Fischer 2013). With regard to dendrochronology analysis, we selected twenty trees within each plot. Only living trees without visible specific symptoms of Armillaria spp. infestation (presence of mushrooms, intensive resinosis on the base of the trunk near the soil line and noticeable hypertrophy of the base of the trunk) or signs of bark beetle attack were assessed. We visually evaluated the assessable crowns using modified methods of Cudlín et al. (2001) and Eichhorn et al. (2010) (Table 3). The assessable crown includes recently dead branches but excludes snags that have been dead for many years (i.e., those that have already lost their side-shoots). For the assessment of parameters of the tree parts that are five or more metres above ground, the use of binoculars (10×50) is mandatory. We also evaluated the stress response of all trees (Cudlín et al., 2001). The stress response classification illustrates the stress tolerance and crown structure transformation stage (i.e., rate of substitution of original primary shoots by secondary shoots). Based on the guidelines of Cudlín et al. (2001), we applied the following classes: A resistant tree has total defoliation lower than 35% and a proportion of secondary shoots lower than 50%; a resilient tree has total defoliation lower than 35% and a proportion of secondary shoots higher than 50%; a slightly transformed damaged tree has total defoliation higher than 35% and a proportion of secondary shoots lower than 50%; a considerably transformed damaged tree has total defoliation higher than 35% and a pro-portion of secondary shoots higher than 50%.

Table 3. The evaluated crown condition parameters (Lesinski & Landmann, 1985; Cudlín et al., 2001; Eichhorn et al., 2010)

Categorical tree crown parameters were compared with the mean tree-ring width index (TRWi) for the last 5, 10 and 15 years using one-way analysis of variance (ANOVA). Semi-continuous tree crown parameters were compared with TRWi for the last 5, 10, and 15 years using Pearson correlation. Although crown parameters indicate the current crown state, they also capture the long-term processes; in particular, the proportion of secondary shoots reflects approximately the last ten years of tree growth.

Tree-ring sampling and chronology development

We used the same trees to analyse tree-ring widths as were used for the crown condition assessment. We used the Pressler borer to extract all samples at breast height. To avoid compression wood, the cores were sampled in a direction parallel to the slope. Because the between-tree variability of TRW within a plot is much higher than the within-tree variability around the stem (Bošeľa et al., 2014), we extracted one core per tree. We measured the samples using the VIAS TimeTable (Vienna Institute for Archaeological Science, Vienna, Austria) measuring system (with an accuracy of 0.01 mm). Then, we synchronized and cross-dated the tree-ring sequences using PAST4 software (©SCIEM) and COFECHA (Grissino-Mayer, 2001). To assess the degree of similarity between the raw TRW series, we used the t-test according to Baillie & Pilcher (1973) and the t-test according to Hollstein (1980), the coefficient of agreement (called the Gleichläufigkeit) (Eckstein & Bauch, 1969), the correlation coefficient, and a visual comparison of the TRW series, which is crucial for the final dating (Rybníček et al., 2007). Well-correlated TRW series were used to create site (plot) TRW chronologies. Non-climatic size- and age-related growth trends and other factors (e.g., competition) were removed from the individual TRW series by applying negative exponential functions (Cook & Peters, 1981) in the ARSTAN software (Cook & Krusic, 2005). This standardization method best resembles the common growth trend and preserves inter-annual to multi-decadal growth variations. TRW indices (TRWi) were calculated as residuals between the measured TRW and the corresponding fitted values after power transformation (Cook & Peters, 1997). The mean chronologies were calculated using bi-weight robust means, and their signal strength was assessed using the inter-series correlation (Rbar) and the expressed population signal (EPS; Wigley et al., 1984). Principal component analysis (PCA) was applied to capture the main growth variance among site-indexed chronologies for the common period (1961–2014).

Growth-climate response

To calculate the correlation between radial incre-ments and climate characteristics (temperature, pre-cipitation, AWR and PDSI), we used the residual indexed tree-ring width chronologies (TRWi) in the DendroClim2002 software (Biondi & Waikul, 2004) for the period 1961–2014. Pearson’s correlation coefficients were calculated for the seasonal window from May of the previous year to August of the year of tree-ring formation – referred to as “the current year”. This interval should have the maximum impact on the TRW in the study area (Gričar et al., 2014). In addition, the seasonal means (temperature, PDSI and AWR) or totals (precipitation) of the previous and current years were analysed. Furthermore, we calculated 20-year backward moving Pearson’s correlations to explore the expected temporal changes. This window length is a compromise between isolating signal changes with the highest possible temporal resolution and having enough data-points to estimate the signal (Friedrichs et al., 2009). The negative pointer years were analysed for the period replicated by at least 20 TRWi series (from 1930 to the present). The negative pointer years were analysed for years in which residual TRWi chronology dropped below -0.5 standard deviation (SD). The threshold value was arbitrarily defined to yield a sufficient number of extreme years. The relationships between negative pointer years and climate factors were determined using logistic regression (Quinn & Keough, 2002), for which the binary response was coded as a “normal” year (value zero) or a negative pointer year (value 1). Models were verified in the first step using Wald’s test for regression parameters and goodness of fit (Quinn & Keough, 2002), and in the second step, using the likelihood ratio test. The model was considered a good fit if both of these tests reached the 0.05 significance level.


Crown condition and its effect on the mean TRW of the last fifteen years

In total, we assessed 120 Norway spruce trees in the study area. The considerably transformed damaged trees predominated (Fig. 3). The average defoliation was 42%, and the average percentage of secondary shoots was 51%. Thirty-one percent of trees had abnormal crown tops (short – retarded 23%, broken 8%) and 11% had abnormal crown shapes (wide 5%, narrow 3%, uneven 3%); 41% of trees had the branch form changed from comb to brush as a consequence of repeated substitution of original primary shoots by secondary shoots. We observed yellowing only in small parts of the crowns (maximum 20% of crown) of 14% of the trees; browning was not present at all. The trees with changed shape and crown tops, as well as discoloured trees, were dispersed throughout the stand, without recognizable grouping or effects of prevailing wind direction.

Figure 3. Crown conditions. The bubble size represents the number of trees with the given values of the total defoliation and the proportion of secondary shoots – this is a continuum from the smallest bubble representing one tree to the largest bubble representing 13 trees.

The last 5-year, 10-year and 15-year tree-ring widths of the trees with abnormal crown tops and abnormal crown shapes were significantly narrower than in the trees with normal crowns (Table 4). The last 5-year tree-ring width of trees with a lower % of secondary shoots was significantly narrower than in the trees with an unchanged branch form (Table 5).

Table 4. Results of the effect of categorical crown condition parameters on the mean tree-ring width (TRWi) of the last 5, 10 and 15 years analysed using one-way ANOVA. Statistically significant values (p < 0.05) are in bold.

Table 5. Results of the effect of semi-continuous crown condition parameters on the mean tree-ring width (TRWi) of the last 5, 10 and 15 years analysed using Pearson correlation. Significant values of the Pearson correlation coefficient are given in bold, significant level: p < 0.05.

Climate-growth relationships

Given that residual site TRWi chronologies showed high common variability (Fig. 4, the PC1 axis explained more than 77% of the variance) and were used to com-pile the regional TRW chronology (Fig. 5A), the re-gional chronology covering the 1915–2014 period was characterized by an average growth rate higher than 2.5 mm and a high first order autocorrelation (0.77). The length of the TRW series ranged from 50 to 100 years with a mean segment length of 75 years. Reliable values of Rbar (mean = 0.4) and EPS (> 0.95) revealed a robust signal strength of the chronology (Fig. 5B). Although the more recent part of the study period (1991–2014) was shorter compared to the normal period (1961–1990), more negative pointer years were detected in the last two decades.

Figure 4. PCA of indexed site TRW chronologies for the common overlap period of 1945–2014 (A) and the scree plot (B).

Figure 5. (A) Raw (black), standard (dark grey) and residual (light grey) versions of the regional TRW chronology, its replication (dashed line) and basic dendrochronological metrics (AGR – average growth rate, SD – standard deviation, AC1 – first-order autocorrelation. Black dots indicate negative pointer years. (B) Rbar (mean inter-series correlation) and EPS (expressed population signal) statistics of the regional TRW chronology.

The months that had significant correlations be-tween TRW and climate parameters in the whole period 1961–2014 are presented (Fig. 6). Tree-ring width was significantly influenced by climate mainly in September and October of the previous year and July and August of the current year. September temperature had a negative effect on TRW. The rest of the observed significant correlations were positive (Fig. 6).

Figure 6. Correlation coefficients calculated between mean monthly temperature, monthly precipitation totals, relative soil water content for the top 1.3 m (AWR), Palmer drought severity index (PDSI) and tree-ring width (TRW) for 20-year backward moving intervals. Significant correlations are indicated by bold lines (Student’s t test). Only months that had significant correlations in the whole period 1961–2014 are shown. The years on the x-axis represent the last year of moving intervals.

The significance of many observed correlations was not stable over time (Fig. 6). Most relationships were significant only in the last two or three decades. Previous September temperature, precipitation, and AWR became significant in moving intervals ending after 1997 (AWR) and 1999 (temperature and precipitation). Previous August and September PDSI were significant at the beginning and end of the analysed period. Previous October AWR and PDSI became significant in moving intervals ending after 1990 and 2000, respectively. The effects of PDSI in the previous November and December became significant at the end of the analysed period. In contrast, the effect of the previous July precipitation lost its significance in intervals after 1979–1999.

The effects of current July precipitation, AWR and PDSI became significant in moving intervals ending after 1997 (precipitation and AWR) and 1999 (PDSI). August AWR and PDSI became significant in moving intervals ending after 2000 and 1999, respectively. The effects of January and February PDSI became significant at the end of the analysed period. The effects of temperature in March and precipitation in August were only significant in scattered short intervals.

The results of the logistic regression showed that negative pointer years were caused by drought in the 1991–2014 period. Low values of AWR and PDSI during the previous and current years of tree-ring formation significantly correlated with the occurrence of negative pointer years. However, no significant relationship between negative pointer years in the first part of the study period (1961–1990) and climate factors was revealed.


Relationships between crown condition and TRWi

The observed mean defoliation in all stands was higher than the mean defoliation in the Czech Republic (MZe, 2016). We propose that the assessed spruces defoliated and transformed their crown structures in reaction to drought periods after 2012 (Fig. 1) because the research area was one of the Czech land areas where the drought periods were strongest (Zahradníček et al., 2013, 2014).

In general, trees have different drought avoidance and drought tolerance given their abilities to modulate their vegetative and reproductive growth according to water availability and their abilities to maintain (re-latively) higher tissue water content despite reduced soil water content. The differences in root morphology, presence of decay, and water conducting element conditions lead to differences in the health status of trees, but not all of the trees show visible symptoms of stress. Defoliation and radial growth can complement each other, but their reactions to stress are usually not synchronous. Drought often affects radial growth almost immediately, while foliage reduction becomes visible months later (Dobbertin, 2006). This reaction asynchronicity could also be the reason we did not identify any relationships between defoliation and the mean TRWi in recent years.

The observed effects of the changed crown shapes and crown tops in the last 5-year, 10-year and 15-year tree-ring widths testified to a longer duration of crown changes. Most of the crown reductions were probably mainly caused by the wind, heavy snow and frost deposits over fifteen years ago. The trees were dispersed throughout the stand, without recognizable effects of prevailing wind direction. On the other hand, the process of crown structure transformation, i.e., the substitution of original primary shoots by secondary shoots probably proceeded in the past few years because the observed percentage of secondary shoots significantly affected only the last 5-year TRWi. We presumed that the crown transformation process was initiated by drought periods after 2012 (see above).

Differences in tree growth response to climate or other environmental factors can also indicate differences in ecotype, provenience or individual predisposition as well as micro-site factors and age-related physiology. The different growth reactivity (growth strategies) can also be grounds for different crown reactivity and resistance to biotic agents. It was found in the declining spruce stands in southeast Norway that trees with dieback symptoms were anatomically predisposed to suffer from hydraulic failure during the drought pe-riod because their growth and resource use strategies were different (Rosner et al., 2016). The trees with a provident strategy have a higher wood density, which increased the safety margin for hydraulic failure in years with normal or high water availability. The trees with prodigal strategy invested in growth and increased conductivity rather than hydraulic safety and benefited from it in years with a high water supply but seemed to suffer during dry periods (Hentschel et al. 2014). On the other hand, Børja et al. (2016) identified disruption in the water transport system in symptomatic trees as well as in spruce trees without visual symptoms, and Čermák et al. (2017) did not identify any significant relationships between crown condition and radial growth in the same area affected by a forest decline episode.

Climate-growth relationships

The study of droughts in the Czech lands demonstrates an increasing long-term dryness in the Czech climate. The period 2004–2012 was the most severely dry part of the period with available climatic data (from 1805). Two similar long periods of drought were also recorded in the 19th century, but the drought severity was lower (Brázdil et al. 2014). Most of the climatic parameters significantly correlated with TRW mainly after the end of the 1990s (Fig. 6), and the relationships probably primarily reflect the higher frequency and duration of drought periods (Fig. 2). During 1961–1990, only one-year drought periods affected trees; after 2002, two longer drought periods limited growth conditions. The drought-affected growing seasons were 1992–1994, 2003–2007 and 2012–2015 in the research area (Fig. 1, PDSI).

Brázdil et al. (2014) demonstrated that the drought episodes in the Czech lands before 1880 may be attributed to a lack of precipitation, whereas the droughts of recent decades (particularly 2004–2012) are more strongly related to high temperatures. The increasing temperature increased the vapour pressure deficit and thereby the transpiration rates. The climatic optimum of Picea abies is characterized by a mean annual temperature of 6°C and a growing-season precipitation total of 490–580 mm (Modrzyński, 2007). Although the growing-season precipitation total remained at the bottom edge of the climatic optimum range in our study plots, the temperature increased above the climatic optimum (Table 2). The current climatic conditions are thus not comfortable for Picea abies in the study area.

These changes in climate conditions can be clearly seen in the climate-growth relationship. The increasing negative significant effect of previous September temperature to TRW (Fig. 6) validates the described temperature effect of drought on our assessed area. The effects of previous September temperature and precipitation are also reflected in AWR and PDSI effects on TRW in the following autumn months (Fig. 6). A negative effect of temperature in the previous autumn was identified many times across Europe (MZe ČR & VÚLHM, 2004; Andreassen et al., 2006; Bouriaud & Popa, 2009; Affolter et al., 2010; Aakala & Kuuluvainen, 2011; Rybníček et al., 2012b, c). However, a positive effect of precipitation in the previous autumn has also been observed (Bouriaud & Popa, 2009; Rybníček et al., 2010). Higher mean temperatures at the end of the growing season generally negatively affect the availability of soil moisture through increased evapotranspiration (Miyamoto et al., 2010). Conversely, higher autumn precipitation increases soil water content and the amount of available water in the spring. Moreover, September is important for the growth of Norway spruce roots because their second distinct growth period is from August to September (Xu et al., 1997).

In the same period when the previous September precipitation started to be significant, the previous July precipitation stopped being significant (Fig. 6). The relationship between previous summer precipitation and TRW is frequent (Affolter et al., 2010; Aakala & Kuuluvainen, 2011), especially in the conditions of Central Europe (Rybníček et al., 2009, 2010, 2012c). The relationship is also commonly explained as a positive effect on soil water content and production capacity (vitality, distribution and biomass of roots, etc.). However, in the above-described climate conditions, the previous July precipitation effect on growing conditions at the beginning of the growing season is probably minimized.

Our results also show that precipitation in the current summer months (mainly in July) and the related AWR and PDSI are significant mainly in moving intervals ending at the end of the 1990s or the beginning of the 2000s (Fig. 6). The significant positive correlation between July or August precipitation and TRW was identified in the central area of the Czech Republic (Rybníček et al., 2012c), the White Carpathians (Kolář et al., 2017), in northern lowland Poland (Koprowski & Zielski, 2006) and the Alps (Affolter et al., 2010). The observed increase in July and August AWR and PDSI significance is probably also related to the increasing temperature. The mean June temperature increased from 13.9°C in 1961–1990 to 14.9°C in 1991–2014, the mean July temperature increased from 15.4°C to 17.2°C, and the mean August temperature increased from 15.2°C to 17°C. The strong temperature growth increased evapotranspiration and led to decreased relative soil water content for the top 1.3 m.

Intensive and often recurring drought spells (Fig. 2) starting at the beginning of the 1990s caused more frequent spruce growth depression reflected in a higher number of negative pointer years compared to the normal period (Fig. 5A). The results of the logistic regression revealed that the negative pointer years were triggered by a lack of available water, but only in recent decades. Therefore, the expected more frequent drought in the following years can lead to serious growth reductions and possible weakening of tree vitality.


The observed crown condition and its relationships with the last 5-year, 10-year and 15-year TRWi indicate stress intensification during the period starting at the turn of the 21st century.

The soil water content decrease evoked mainly by increasing temperature led to significant changes in relationships between climate parameters and tree-ring width. We suppose that water availability was the main inciting factor (Sinclair, 1967) that triggered the decline event and is currently a contributing factor that significantly intensifies forest decline.

The study area was relatively favourable for Picea abies at the time when the stands were planted (from 1915 to 1965), but currently, the climatic conditions at these sites have changed, moving away from the Norway spruce ecological optimum (Modrzyński, 2007). Silviculture of the Norway spruce is extremely risky in such conditions. The conversion from Norway spruce forests managed under systems involving coupes to uneven-aged mixed forest silviculture is needed under such conditions (Temperli et al., 2012). Long-term general forestry planning uses time horizons of 20 years or longer. Our results have shown that the observed climate change is too dynamic for many recommendations included in the plans for the site and climate conditions – recommended species composition, timber-stand improvement models, etc. The recommendations become inadequate in the course of the planning period, let alone during the tree or stand life. New long-term plans should support operational decisions responding to the state of the stands. It is necessary for effective adaptive management.


Aakala T, Kuuluvainen T, 2011. Summer droughts depress ra-dial growth of Picea abies in pristine taiga of the Arkhan-gelsk province, northwestern Russia. Dendrochronologia 29: 67-75.

Affolter P, Büntgen U, Esper J, Rigling A, Weber P, Luterbacher J, Frank D, 2010. Inner Alpine conifer res-ponse to 20th century drought swings. Eur J For Res 129: 289-298.

Allen CD, Macalady AK, Chenchouni H, Bachelet D, Mcdowell N, Vennetier M, Kitzberger T, Rigling A, Breshears DD, Hogg EH, et al., 2010. A global overview of drought and heat-induced tree mortality reveals emerging climate change risks for forests. Forest Ecol Manag 259(4): 660-684.

Andreassen K, Solberg S, Tveito OE, Lystad SL, 2006. Regional differences in climatic responses of Norway spruce (Picea abies L. Karst) growth in Norway. Forest Ecol Manag 222: 211-221.

Baillie MGL, Pilcher JR, 1973. A simple crossdating program for tree-ring research. Tree-Ring Bull 33:7-14.

Biondi F, Waikul K, 2004. DendroClim2002: AC++programfor statistical calibration of climate signals in tree ring chronologies. Comput Geosci 30: 303-311.

Børja I, Svĕtlík J, Nadezhdin V, Čermák J, Rosner S, Nadezhdina N, 2016. Sap flux - a real time assessment of health status in Norway spruce. Scand J For Res 31(5): 450-457.

Bošeľa M, Sedmák R, MaruŠák R, Sedmáková D, Petráš R, Barna M, 2014. Evaluating similarity of radial increments around tree stem circumference of European beech and Norway spruce from Central Europe. Geochronometria 41(2): 136-146.

Bouriaud O, Popa I, 2009. Comparative dendroclimatic study of Scots pine, Norway spruce, and silver fir in the Vrancea Range, Eastern Carpathian Mountains. Trees 23: 95-106.

Brázdil R, Trnka M, Mikšovský J, Řezníčková L, Dobrovolný P, 2014. Spring‐summer droughts in the Czech Land in 1805-2012 and their forcings. Int J Climatol 35: 1405-1421.

Camarero JJ, Gazol A, Sangüesa-BarredaG, Cantero A, Sánchez-Salguero R, Sánchez-Miranda A, Granda E, Serra-Maluquer X, Ibáñez R, 2018. Forest Growth Responses to Drought at Short- and Long-Term Scales in Spain: Squeezing the Stress Memory from Tree Rings. Front Ecol Evol 6, 9.

Čermák P, Rybníček M, Žid T, Andreasssen K, Børja I, Kolář T, 2017. Impact of climate change on growth dynamics of Norway spruce in south-eastern Norway. Silva Fenn 51(2).

Cienciala E, Tumajer J, Zatloukal V, Beranová J, Holá Š, Hůnová I, Russ R, 2017. Recent spruce decline with biotic pathogen infestation as a result of interacting climate, deposition and soil variables. Eur J For Res 136: 307-317.

Cook ER, Krusic PJ, 2005. ARSTAN v. 41d: a Tree-ring Standardization ProgramBased on Detrending and Autoregressive Time Series Modeling, with Interactive Graphics. New York: Tree-Ring Laboratory, Lamont-Doherty Earth Observatory of Columbia University, Palisades.

Cook ER, Peters K, 1981. The smoothing spline: A new approach to standardizing forest interior tree-ring width series for dendroclimatic studies. Tree-Ring Bull 41: 45-53.

Cook ER, Peters K, 1997. Calculating unbiased tree-ring indices for the study of climatic and envi­ronmental change. Holocene 7: 361-370.

Cudlín P, Novotný R, Moravec I, Chmelíková E, 2001. Retrospective evaluation of the response of montane forest ecosystems to multiple stress. Ekológia 20: 108-124.

Dobbertin M, 2006. Tree growth as indicator of tree vitality and of tree reaction to environmental stress: A review. Eur J For Res 124(4): 319-333.

Eckstein D, Bauch J, 1969. Beitrag zur Rationalisierung eines dendrochronologischen Verfahrens und zur Analyse seiner Aussagesicherheit. Cent bl gesamte Forstwes 88: 230-250.

Eichhorn J, Roskams P, Ferretti M, Mues V, Szepesi A, Durrant D, 2010. Visual Assessment of Crown Condition and Damaging Agents. Manual Part IV. In: Manual on methods and criteria for harmonized sampling, assessment, monitoring and analysis of the effects of air pollution on forests. UNECE ICP Forests Programme Co-ordinating Centre, Hamburg. [cited 2017 November 20] Available from:

Ferretti M., Fischer R. (eds.) 2013. Forest Monitoring. Methods for terrestrial investigations in Europe with an overview of North America and Asia. Elsevier Ltd., Oxford. 507 pp.

Franklin JF, Shugart HH, Harmon ME, 1987. Tree death as an ecological process. Bioscience 27: 550-556.

Friedrichs DA, Trouet V, Büntgen U, Frank DC, Esper J, Neuwirth B, Löffler J, 2009. Species-specific climate sensitivity of tree growth in central-west Germany. Trees - Struct Funct 23: 729-739.

Fritts HC (1976) Tree rings and climate. Academic Press, London. 582 pp.

Gazol A, Camarero JJ, Anderegg, WRL, Vicente-Serrano SM, 2017. Impacts of droughts on the growth resilience of Northern Hemisphere forests. Glob. Ecol. Biogeogr. 26(2): 166-176.

Greenwood S, Ruiz-Benito P, Martínez-Vilalta J, Lioret F, Kitzberge, T, Alle, CD, Fensham R, Laughlin DC, Kattge J, Bönisch G, et al. 2017. Tree mortality across biomes is promoted by drought intensity, lower wood density and higher specific leaf area. Ecol Lett 20(4): 539-553.

Gričar J, Prislan P, Gryc V, Vavrčík H, de Luis M, Čufar K, 2014. Plastic and locally adapted phenology in cambial seasonality and production of xylem and phloem cells in Picea abies from temperate environments. Tree Physiol 34(8): 869-881.

Grissino-Mayer HD, 2001. Evaluating crossdating accuracy: A manual and tutorial for the computer program COFECHA. Tree-Ring Res 57 (2): 205-221.

Grodzki W, 2007. Spatio-temporal patterns of the Norway spruce decline in the Beskid Śląski and Żywiecki (Western Carpathians) in southern Poland. J For Sci 53(Special Issue): 38-44.

Grodzki W, 2010. The decline of Norway spruce Picea abies (L.) Karst. stands in Beskid Śląski and Źywiecki: theoretical concept and reality. Beskydy 3: 19-26.

Hartl-Meier C, Zang C, Büntgen U, Esper J, Rothe A, Göttlein A, Dirnböck T, Treydte K, 2014. Uniform climate sensitivity in tree-ring stable isotopes across species and sites in a mid-latitude temperate forest. Tree Physiol 35, 4-15

Hentschel R, Rosner S, Kayler ZE, Andreassen K, Borja I, Solberg S, Tveito OE, Priesack E, Gessler A, 2014. Norway spruce physiological and anatomical predisposition to dieback. For Ecol Manag 322: 27-36.

Hlavinka P, Trnka M, Balek J, 2011. Development and evaluation of the SoilClimmodel for water balance and soil climate estimates. Agric. Water Manag 98(8): 1249-1261.

Hollstein E, 1980. Mitteleuropäische Eichenchronologie. Triererdendrochronologische Forschungen zur Archäo-logie und Kunstgeschichte. Trierer Grabungen und For-schungen. Mainz am Rhein. 274 pp.

Holuša J, Lubojacký J, Čurn V, Tonka T, Lukášová K, Horák J, 2018. Combined effects of drought stress and Armillaria infection on tree mortality in Norway spruce plantations. For Ecol Manag 427(1): 434-445.

Jankovský L, Cudlín P, Moravec I, 2003. Root decays as a potential predisposition factor of a bark beetle disaster in the Šumava Mts. J For Sci 49(3): 125-132.

Kolář T, Čermák P, Trnka M, Žid T, Rybníček M, 2017. Temporal changes in the climate sensitivity of Norway spruce and European beech along an elevation gradient in Central Europe. Agric For Meteorol 239: 24-33.

Koprowski M, Zielski A, 2006. Dendrochronology of Norway spruce (Picea abies (L.) Karst.) from two range centres in lowland Poland. Trees 20: 383-390.

Lesinski J, Landmann G, 1985. Crown and branch mal-formation in confers related to forest decline. In: Cape, J.N., et al. (eds.) Scientific basis of forest decline symptomatology, Brussels: Commision of European Communities, Air Pollution Research Report 15: 95-105.

Leuschner C, Ellenberg, H, 2017. Ecology of Central European Forests. Vegetation Ecology of Central Europe, Volume I. Springer International Publishing. 1st edition. 972 pp.

Manion PD, 1991. Tree disease concepts. Prentice-Hall Inc, Engelwood Clifts, N.J., 402 pp.

Manion PD, Lachance D (eds.), 1992. Forest decline concepts. APS Press, St. Paul, MN, 249 pp.

Miyamoto Y, Griesbauer HP, Green DS, 2010. Growth responses of tree coexisting conifer species to climate wide geographic and climate ranges in Yukon and British Columbia. For Ecol Manag 259: 514-523.

Modrzyński J, 2007. Outline of Ecology in Tjoelker, MG, Borytański A, Bugała W, 2007. Biology and Ecology of Norway Spruce. Springer, Dordreeht. 195-198.

MZe ČR, VÚLHM, 2004. Monitoring stavu lesa v České republice: 1984-2003 [Forest condition monitoring in the Czech Republic: ICP Forests] MZe ČR, Praha and VÚLHM, Jíloviště-Strnady. 431 pp.

MZe ČR, 2016. Zpráva o stavu lesa a lesního hospodářství České republiky v roce 2015. [Information on Forests and Forestry in the Czech Republic by 2015] MZe ČR, Praha. 134 pp.

Novotný R, Lachmanová Z, Šrámek V, Vortelová L, 2008. Air pollution load and stand nutrition in the Forest District Jablunkov, part Nýdek. J For Sci 54(2): 49-54.

Palmer WC, 1965. Meteorological Drought. Research Paper, US Weather Bureau, Washington, DC, 45: 58.

Quinn GP, Keough MJ. 2002. Experimental Designs and Data Analysis for Biologists. Cambridge University Press, Cambridge. 537 pp.

Rosner S, Svĕtlík J, Andreassen K, Børja I, Dalsgaard L, Evans R, Luss S., Tveito OE, Solberg S, 2016. Novel hydraulic vulnerability proxies for boreal conifer species reveal that opportunists may have lower survival prospects under extreme climatic events. Front Plant Sci 7 (831).

Rybníček M, Čermák P, Hadaš P, Kolář T, Žid, 2012a. Dendrochronological analysis and habitual stress dia-gnostic assessment of Norway spruce (Picea abies) stands in the Drahany Highlands. Wood Res 57(2): 189-206.

Rybníček M, Čermák P, Kolář T, Přemyslovská E, Žid T, 2009. Influence of temperatures and precipitation on radial increment of Orlické hory Mts. spruce stands at altitudes over 800 m a.s.l. J For Sci 55(6): 257-263.

Rybníček M, Čermák P, Kolář T, Žid T, 2012b. Growth responses of Norway spruce (Picea abies) to the climate in the south-eastern part of the Českomoravská Upland (Czech Republic). Geochronometria 39(2): 149-157.

Rybníček M, Čermák P, Žid T, Kolář T, 2010. Radial growth and health conditions of Norway spruce (Picea abies (L.) Karst.) stands in relation to climate (Silesian Beskids, Czech Republic). Geochronometria 36: 9-16.

Rybníček M, Čermák P, Žid T, Kolář T, 2012c. Growth responses of Picea abies to climate in the central part of the Českomoravská Upland (Czech Republic). Dendrobiology 68: 21-30.

Rybníček M, Gryc V, Vavrčík H, Horáček P, 2007. Annual ring analysis of the root system of Scots pine. Wood Res 52 (3): 1-14.

Sinclair WA, 1967. Decline of hardwoods: possible causes. Proceedings of the International Shade Tree Conference 43: 17-32.

Šrámek V, Novotný R, Fadrhonsová V, 2015. Decay of Norway spruce stands and quality of forest soil in the region of northern Moravia and Silesia. ZLV 60(2): 147-153.

Šrámek V, Vejpustková M, Novotný R, Hellebrandová K, 2008. Yellowing of Norway spruce stands in the Silesian Beskids - damage extent and dynamics. J For Sc. 54(2): 55-63.

Štěpánek P, Zahradníček P, Skalák P, 2009. Data quality control and homogenization of air temperature and precipitation series in the area of the Czech Republic in the period 1961-2007. ASR 3: 23-26.

Štěpánek P, Zahradníček P, Huth R, 2011. Interpolation techniques used for data quality control and calculation of technical series: An example of Central European daily time series. Idöjárás 115 (1-2): 87-98.

Sturrock RN, Frankel SJ, Brown AW, Hennon PE, Kliejunas J, Lewis KJ, Worrall JJ, Woods AJ, 2011. Climate change and forest diseases. Plant Pathol 60: 133-149.

Temperli C, Bugmann H, Elkin C, 2012. Adaptive ma-nagement for competing forest goods and services under climate change. Ecol Appl 22(8): 2065-2077.

Trumbore S, Brando P, Hartmann, H, 2015. Forest health and global change. Science 349(6250): 814-818.

Vakula J, Zúbrik M, Galko J, Gubka A, Kunca A, Nikolov C, Bošela M, 2015. Influence of selected factors on bark beetle outbreak dynamics in the Western Carpathians. Cent Eur For J 61: 149-156.

van der Maaten-Theunissen M, van der Maaten E, Kahle H-P, 2013. Drought sensitivity of Norway spruce is higher than that of silver fir along an altitudinal gradient in southwestern Germany. Ann For Sci 70(2):185-193.

Viewegh J, Kusbach A, Mikeska M, 2003. Czech forest ecosystem classification. J For Sci 49: 74-82.

Wigley TML, Briffa KR, Jones PD, 1984. On the average value of correlated time series, with applications in dendroclimatology and hydrometeorology. J. Appl Meteorol Climatol 23: 201-213.

Xu Z, Röhrig E, Fölster H, 1997. Reaction of root systems of grand fir (Abies grandis Lindl.) and Norway spruce (Picea abies Karst.) to seasonal waterlogging. For Ecol Manag 93: 9-19.

Zahradníček P, Trnka M, Brázdil R, Možný M, Štěpánek P, 2013. Extreme drought episode from August 2011 to May 2012 in the Czech Lands. EGU General Assembly 2013, held 7-12 April, 2013 in Vienna, Austria, p.13137.

Zahradníček P, Trnka M, Štěpánek P., Semeradova D, Farda A, 2014. Drought periods in 2014 In: Rožnovský, J., Li-tschmann, T., (eds): Mendel a bioklimatologie. Brno, 3.-5. 9.