Floodplain forests under stress: how ash dieback and hydrology affect tree growth patterns under climate change

doi:10.21203/rs.3.rs-4676274/v1

Download PDF

Article

Floodplain forests under stress: how ash dieback and hydrology affect tree growth patterns under climate change

https://doi.org/10.21203/rs.3.rs-4676274/v1

This work is licensed under a CC BY 4.0 License

You are reading this latest preprint version

Floodplain forests are currently undergoing substantial reorganization processes due to the combined effects of management-induced altered hydrological conditions, climate change and novel invasive pathogens. Nowadays, the ash dieback is one of the most concerning diseases affecting temperate floodplain forests, causing substantial tree mortality and threatening the loss of the dominant key tree species of the hardwood floodplain forest, Fraxinus excelsior. Understanding how the increased light availability caused by pathogen-driven mortality in combination with altered hydrological conditions and climate change affects growth responses in a diverse forest community is of crucial importance for conservation efforts. Thus, we examined growth of the main tree species in response to ash dieback and how it depended on altered hydrological conditions under novel climatic conditions for the lower and upper canopy in the floodplain forest of Leipzig, Germany. Our study period encompassed the consecutive drought years from 2018 to 2020. We found that tree growth responded mostly positively to increased light availability, but only on moist sites, while tree growth largely declined on dry sites, suggesting that water availability is a critical factor for tree species to be able to benefit from increased light availability due to canopy disturbances caused by ash dieback. This hydrological effect was species specific in the lower canopy but not in the upper canopy. While, in the lower canopy, some species such as the competitive shade-tolerant but flood-intolerant Acer pseudoplatanus and Acer platanoides benefited from ash dieback on moist sites, others were less affected or suffered disproportionally, indicating that floodplain forests might turn into a novel ecosystem dominated by competitive Acer species, which may have detrimental effects on ecosystem functioning. Our results give hints on floodplain forests of the future and have important implications for conservation measures, suggesting that a substantial revitalization of natural hydrological dynamics is important to maintain a tree composition that resembles the existing one and thus sustain their conservation status.

Biological sciences/Ecology/Biodiversity

Biological sciences/Ecology/Forest ecology

Floodplain forest

tree growth

ash dieback

climate change

hydrology

forest reorganization

Forest ecosystems around the globe are currently undergoing complex reorganization processes, as they are facing increasing natural and anthropogenic disturbances amplified by continuing climate change ^1,2. Disturbances and novel stressors, such as invasive pathogens, wildfires, windthrows, droughts and heat waves, are important drivers of forest ecosystem dynamics and can cause substantial tree mortality and trigger abrupt changes in growing conditions ^1,3–5. The initial years following a disturbance, the reorganization phase, is crucial, as early colonizers and the dynamics of advanced regeneration in the understory often shape the forest's structure and composition for decades and centuries to come ^1,6,7. Under climate change and anthropogenic influence, these reorganization processes fundamentally deviate from well-known succession dynamics as disturbances nowadays coincide with significantly altered conditions, potentially catalyzing the system’s shift into a new regime and hampering the forests’ provision of ecosystem services ^2,8. Floodplain forests are a prime example of ecosystems currently undergoing significant reorganization processes due to multiple coinciding compound effects of global change. They are also one of the most dynamic, productive and diverse, yet one of the most endangered ecosystems in the world ^9–13. Understanding these novel reorganization processes and post-disturbance growth responses under climate change is of crucial importance for conservation efforts which is why we examine the effect of a novel pathogen under climate change on the growth dynamics of a species-rich and protected floodplain forest.

The stressors floodplain forests currently experience are manyfold. First, extensive hydraulic engineering measures of the past such as river straightening and dike constructions led to the absence of floods and sinking ground water levels, causing floodplains across Central Europe to dry up on a large scale ^14–16. Drainage, coupled with altered silvicultural practices shifting away from coppice-with-standards management, allowed untypical floodplain forest species such as the flood-intolerant but shade-tolerant sycamore maple (Acer pseudoplatanus L.) to thrive, while keystone species like the light-demanding pedunculate oak (Quercus robur L.) are quickly vanishing ^17,18. Second, climate change is associated with increases in the frequency, duration, and/or intensity of drought and heat stress which could fundamentally alter the composition and structure of floodplain forests and other forest types across the world, potentially inducing increased tree mortality ^{3,IPCC 19}. From 2018 to 2020, Central Europe was faced with the hottest and driest consecutive years in 250 years of climate records, showing an unprecedented level of intensity ^20,21 which is likely to occur more often as climate change exacerbates ²². Normally, most floodplain species are adapted to surviving dry summer periods, by either accessing groundwater with their roots or by restricting their water consumption ¹⁵. However, several studies have found temperate floodplain forests to be susceptible to drought-induced stress on tree growth ^23–25, and prolonged or consecutive droughts appear to bring floodplain forests closer to a tipping point ^21,24. Third, invasive pathogens are causing a high degree of tree mortality and changing forest structure, with climate change driving these processes ²⁶. Due to warmer climatic conditions, winter mortality in biotic agents of forest pests is likely to be reduced, leading to an increase in outbreaks ²⁷. After the Dutch elm disease spread widely in the last century and has eliminated the majority of elm trees – being a typical floodplain forest species - from the overstory ^17,28, the most concerning disease affecting temperate floodplain systems today is the ash dieback, threatening the loss of another key tree species of the hardwood floodplain forest. Many European floodplain forests have a high percentage of ash in the canopy cover ^29,30, and the imminent collapse of ash populations has far-reaching consequences for biodiversity and the functionality or conservation status of the hardwood floodplain forest ^12,31. Together, these stressors impose critical threats to forest ecosystem functioning.

Tree growth responses to canopy disturbances have been found to be complex and depend on various environmental factors as well as species specific and individual tree characteristics. When a canopy gap forms, the light availability in the canopy increases, but also other abiotic factors, such as temperature, soil moisture and available nutrients as well as the below-ground competitive environment are significantly altered ^32,33. In general, mature trees can respond to local increases in light and below-ground resources depending on their position within the tree layers, their species characteristics, including shade tolerance and water-use strategies, and the related growth conditions prior to the disturbance ³⁴. Diameter growth has been reported to increase in those residual trees, even though the response might be delayed due to the rapid change of growth conditions, which can even cause a temporary decline of growth ^34–38. Smaller overstory trees may benefit more from the relatively higher increase in access to light and thus show stronger growth responses as compared to larger trees that are less limited by competition for resources ^{34,35,37,38,40}. Larger trees may furthermore be less capable of responding to the increased resource availability due to physical constraints associated with age, resulting in limited growth response ^34,41. On the contrary, larger trees may also show an increased growth response and fill the newly available growing space rapidly by extending their crowns ^34,37,42. To better understand growth dynamics of different tree layers in compositionally and structurally complex forests, this study aims at disentangling how lower and upper canopy trees react to canopy disturbances caused by ash dieback.

Species-specific growth responses to canopy disturbances and water availability can be linked to species-specific ecological attributes, encompassing factors such as shade tolerance, drought resistance, and specific demands for water and light. Most studies examining growth responses related to shade tolerance focus on individuals in the juvenile stage ^43,44, whereas little research was done on mature individuals, and results here are ambiguous. Shade-intolerant species were found to react stronger to canopy disturbance ⁴⁵, whereas other studies found shade-tolerant species, especially the smaller individuals, to respond strongly to canopy openings ³⁵. Drought resistance might affect growth responses under altered hydrological conditions, with drought resistant species coping better on dry sites as compared to drought sensitive species ⁴⁶.

In this study, we aimed at investigating one key aspect of early reorganization processes in response to a novel natural disturbance under novel climatic conditions in a diverse European floodplain forest system. Specifically, we examined growth responses of the main tree species to canopy disturbances and thus increased light availability caused by a fungal pathogen, the ash dieback, in dependence on altered hydrological conditions in the floodplain forest of Leipzig, Saxony, Germany. Due to their high tree species richness ⁴⁷, floodplain forests are ideal for comparative studies of tree species reactions to disturbances depending on water availability as they are among the few ecosystems with coexisting mature trees showing a wide range of water-use strategies ^24,48. We hypothesized that

1.) all species in general show positive growth responses to increased light availability due to ash dieback,

2.) the effect of ash dieback on growth responses is dependent on hydrological conditions, with more pronounced positive effects on moist sites as compared to dry sites,

3.) trees in different canopy strata react in a different way: a) trees in the lower canopy particularly benefit from canopy disturbance as the relative increase in access to light is higher than in upper tree layers, b) trees in the upper canopy show weak responses to ash dieback as they are less likely to be limited by competition for resources, and

4.) growth responses are species specific and we will discuss these species-specific differences in the context of shade tolerance and water demands.

2.1. Study site & design

In this study, we used data collected in the northwestern part of the floodplain forest in Leipzig, Saxony, Germany, within the scope of the project ‘Lebendige Luppe’ ⁴⁹. The Leipzig floodplain forest is one of the largest floodplain forests in Central Europe ^10,50 and highly protected due to its relevance as biodiversity hotspot ^14,21,51. The prevailing climate is continental with an average annual temperature of 9.7°C and a mean annual precipitation of 520 mm (1980–2020; DWD Climate Data Center [CDC], Station Leipzig/Halle, ID 2932). It is situated along the rivers Weiße Elster, Pleiße, Luppe and Parthe and has been shaped by human interventions over the last centuries ⁵². Extensive hydraulic engineering schemes such as dike construction and river regulations led to an altered hydrological regime of the floodplain forest, with a lack of regular flooding for over 70 years ^10,53. The dominant soil type can be characterized as a loamy Vega that developed on a 2–4 m thick layer of clay deposited after upstream soil erosion since 8000 b.c. on quarternary alluvials sediments dominated by gravel and sand ^10,54. The consecutive drought years 2018, 2019 and 2020 have created a large deficit in the plant-available water in the soils of Saxony, which affected the ecosystem of the floodplain forest of Leipzig particularly severely, as the regional water balance was already disrupted by flood prevention measures and a lowered ground water table ^31,53. This prolonged drought appears to have crossed a tipping point for the Leipzig hardwood floodplain forest and may have accelerated the spread and lethal damage of tree diseases ^21,24.

The study design of the ‘Lebendige Luppe’ project encompasses 60 permanent observation plots each 0.25ha in size which cover a gradient in topographic distances to the groundwater level ⁵⁵. We used a stratified sampling design with three strata of distance to groundwater: dry (> 2 m), intermediate (1–2 m) and moist (≤ 1 m) plots to represent the entire hydrological gradient of the floodplain, with 20 plots per stratum. During the course of the project, higher resolution hydrological data loggers were installed allowing the calculation of tree-based groundwater to surface distances using a coupled hydrological model ^{56,57, see 3.2}.. The study site misses regular flooding since 1954 due to flood control measures (dikes, river-straightening etc.), except for winter 2011 and summer 2013, when the area of Leipzig experienced extreme flood events. All plots are located in areas classified as FFH habitat type hardwood floodplain forest (LRT 91F0*) or as starmason-oak-hornbeam-forest (LRT 9160) or other forest stands with an age of ≥ 80 years. F. excelsior, Q. robur and Ulmus spec. are the characteristic hardwood floodplain forest species and crucial for its conservation status as FFH habitat type, whereas Acer species, above all A. pseudoplatanus, are untypical for this forest type and were historically rare or absent ^21,58,59. F. excelsior is by far the most dominant species, thus making its loss through ash dieback detrimental for forest composition and ecosystem functioning.

2.2. Hydrological data

Hydrological conditions were continuously measured on 34 of the 60 plots via groundwater measuring stations reaching 4m into the ground ⁵⁵. The plots for the installation of the groundwater monitoring stations were selected to include a gradient of the surface to groundwater distance. The groundwater conditions are mostly confined or at least semiconfined by a 2.5–3.5 m thick low permeable alluvial clay layer ⁵⁴. The measuring points are groundwater observation wells consisting of HDPE pipes (DN 32/25 mm) with a filter section at the level of the aquifer that was sealed with Bentonite granules. Pressure loggers were installed in the wells to measure the groundwater level (groundwater potential) and temperature in high temporal resolution (TD-Diver Schlumberger Water Services, half-hourly measuring interval). The water levels were converted to absolute heights above sea level using precise D-GPS measurements. Data from > 50 groundwater observation wells were used for calibrating a coupled groundwater surface water model with daily means for a 30m grid of the research area. We calculated tree-based distances to groundwater by subtracting the modeled groundwater surface (based on a coupled groundwater surface water model) ⁵⁶ from the ground level elevation for each individual tree.

2.3. Tree data

Two forest inventories were carried out which monitored all living tree and shrub individuals with a diameter at breast height (DBH) of ≥ 5cm on the whole plot area ^60,61. The first inventory took place in two time periods: for the initial 31 plots in the winter of 2013/14, and after the plot network was extended, the remaining 29 plots were inventoried in the winter of 2016/17. The second inventory took place in the winter of 2020/21 for all plots. Thus, our study period encompassed the severe consecutive drought years from 2018 to 2020, allowing us to study indirect effects of climate change on growth responses of the main tree species.

The exact position of each individual was measured using a tachymeter and georeferenced using a Differential GPS; thus, the distance to groundwater could be determined for each individual tree via the ground level elevation (see 3.2.). Furthermore, for each individual, the total height, the height of the crown base and the DBH was measured, the tree species was determined and its social class (1–6 according to Kraft’s classification ⁶²; 1 = predominant, 2 = dominant, 3 = co-dominant, 4 = dominated, 5 = sheltered, 6 = remnants), the (degree of) canopy cover (fully covered, partially covered, uncovered), the vitality status (vital, normal, suffering) and the degree (1–3) and type of damage (stem damage, crown breakage, dry parts in the crown, visible stem rot, fruit bodies of tree fungi at the trunk, other fungi or damage) was assessed. All trees were tagged with a numbered plaque for better identification in the field ⁶¹.

2.4. Data on ash dieback

In order to monitor the extent and development of ash dieback in the Leipzig floodplain forests, annual evaluations of all ash trees (n = 1075) on the 60 plots are carried out since 2016 ⁶³. We used a standardized 6-step classification method (from damage class 0 = healthy to damage class 5 = dead or dying) to document the symptoms of the infestation ^55,64–67. For the following analyses, we used the data of the ash-dieback assessment of 2017 and 2020.

Ash dieback is caused by the fungus Hymenoscyphus fraxineus and the associated secondary fruit form Chalara fraxinea ^29,30,68. In Europe, the ash dieback was first observed in Poland in 1992 and has since then spread to 22 European countries ⁶⁹; in the Leipzig floodplain forest, it was first documented around 2009 ^21,70. The symptoms of the ash dieback can be first observed in young ash stands through discolouration of the shoots, as well as through wilting, leaf spots, and bark necrosis ⁶⁵. For mature trees, relevant characteristics of the ash dieback are loss of leaves, dead twigs or branches, and in some cases the formation of a secondary crown ⁶⁵. The terminal stages of the disease lead to excessive foliage loss and ultimately to mortality ^65,67, creating gaps in the forest canopy. By 2020, 71% of all assessed ashes in the floodplain forest of Leipzig showed intermediate to strong damage symptoms, while not a single assessed individual was without symptoms ^21,63,70.

2.5. Data Analysis

2.5.1. Data preparation

Prior to data analysis, we checked the data for plausibility and removed cases with inconsistencies regarding the DBH measurements, such as missing measurements or deviations from the standardized measuring height of 130 cm by at least 10 cm (44 out of 8624 total individuals). For the final analysis, we only considered the main tree species with the highest abundances in the Leipzig floodplain forest, namely A. campestre, A. platanoides, A. pseudoplatanus, C. betulus, Q. robur, T. cordata and Ulmus spec. (including U. minor, U. glabra and U. x hollandica), and only included tree individuals that were present during both the first and the second inventory to be able to calculate relative growth rates.

To assess the effect of ash dieback and hydrological conditions on growth responses for different canopy layers (see hypothesis 3), we categorized trees into upper and lower canopy individuals according to their position in the stand's social structure. The position of a tree in the stand and, thus, its growth potential is best described by its social class ⁶². Therefore, we combined Kraft's classes 1–3 (predominant, dominant, co-dominant) to ‘upper canopy’ and classes 4–5 (dominated and sheltered) to ‘lower canopy’.

To assess the damage degree of individual trees, we summed up the degree of all eight damage types per tree individual and inventory, and calculated the difference between damage degrees of the second and first inventory (Δ damage degree) to get an index of change in tree individual health conditions.

We calculated plot density as the total basal area (BA) of all trees per plot (including ash trees and shrubs). Basal area of an individual tree was calculated as

$$\:BA={\left(\frac{DBH}{2}\right)}^{2}*\:{\pi\:}$$

In order to approximate the magnitude of the change in light availability due to ash dieback, we calculated the effective basal area change for all dominant ash trees per plot. To this end, we used the range of defoliation for each damage category in the 6-step classification according to Peters et al. ^66,67 to calculate the mean defoliation difference from 2020 to 2017 as an indicator for the change in light conditions. We multiplied the defoliation difference with the ash basal area of the first inventory to calculate the effective ash basal area change and summed it up per plot. We assume that the higher the effective ash basal area change (hereafter ash dieback intensity), the more light became available on the plots from the first to the second inventory.

As a proxy for water availability, we used the median distance to ground water for the summer months from beginning of May until the end of October over the years from 2014 to 2020 for each individual tree. We excluded the year 2013 as in this year there was a severe flooding in Leipzig which biased the mean values for the following years.

We calculated the relative growth rate of each individual tree as

$$\:RGR=\frac{\left(\:\frac{\left(BA2\:-BA1\right)}{\varDelta\:T}\right)}{BA1}$$

with BA1 being the basal area in the first inventory, BA2 the basal area in the second inventory, and ΔT the time interval between the two inventories.

2.5.2. Statistical Analysis

We used Linear Mixed Models (LMMs) ⁷¹ to estimate the effect of canopy disturbance due to the ash dieback and hydrological conditions on growth responses of the main tree species in the Leipzig floodplain forest. We furthermore expected different growth reactions in the lower canopy (LC) and in the upper canopy (UC). To avoid model stability issues and hardly interpretable results due to an unbalanced dataset in terms of species and canopy layer distribution, we ran two separate models for LC and UC trees, respectively, and excluded Q. robur in the LC model and the small-statured A. campestre and Ulmus spec. in the UC model as there are hardly any oaks in the understorey and hardly any individuals of A. campestre and Ulmus spec. in the overstorey (see Table 1). Model structure was otherwise identical for both models. In both models, we used the relative growth rate as response variable and included species, effective ash basal area change (hereafter ash dieback) and distance to ground water as fixed effects (test predictors). We also included all interactions up to order three between all test predictors as we expected species-specific growth responses in dependence on the different environmental factors. To control for the effect of resource competition on growth responses, we included plot density as a fixed effect. To control for individual tree growth conditions, we also included Δ damage degree and the vitality state of each tree in the second inventory as fixed effects. To control for the different timespans between the first and second inventory, we furthermore included the time difference in years between the first and second inventory (ΔT) as fixed effect. We included plot ID as random intercepts effect to control for plot specific effects. Furthermore, we had 129 trees that had more than one stem below breast height of 130 cm (“Zwiesel”). In order to control for non-independent growth responses of these bifurcated trees, we included tree ID as an additional random intercepts effect. Tree ID is defined as the tree including all of its forks. To avoid the model being overconfident with regard to the precision of fixed effects estimates and to keep type I error rate at the nominal level of 5%, we included all theoretically identifiable random slopes ^72,73. More precisely, we included random slopes of species, distance to ground water, delta damage degree and vitality within plot ID (species and vitality were manually dummy coded and then centered before entering them into the random effects part). We also included parameters for the correlations among random intercepts and slopes.

Table 1

Numbers of individuals per species that were present in both the first and second inventory for the lower and upper canopy, respectively.
species	lower canopy	upper canopy
Acer campestre	368	21
Acer platanoides	392	43
Acer pseudoplatanus	786	270
Carpinus betulus	475	74
Quercus robur	26	311
Tilia cordata	494	153
Ulmus spec.	1070	26
total	3611	898

As an overall test of the effects of the test predictors and their interactions, we conducted a full-null model comparison, aiming at avoiding cryptic multiple testing ⁷⁴, whereby the null model lacked the test predictors and their interactions in the fixed effects part but was otherwise identical to the full model. This comparison was based on a likelihood ratio test (LRT) ⁷⁵. We tested the effect of individual fixed effects by means of the Satterthwaite approximation ⁷⁶ using the function lmer of the package lmerTest (version 3.1.3) ⁷⁷ and a model fitted with restricted maximum likelihood.

Prior to fitting the model, we inspected all quantitative predictors and the response for whether their distributions were roughly symmetrical. As a consequence, we log-transformed RGR (after adding 0.01). We only used values of RGR > 0. We then z-transformed ash dieback, plot density, distance to ground water, Δ damage degree and ΔT to achieve an easier interpretable model ⁷⁸ and ease model convergence.

After fitting the model, we checked whether the assumptions of normally distributed and homogeneous residuals were fulfilled by visual inspection of a QQ-plot of residuals ⁷⁹ and residuals plotted against fitted values ⁸⁰. These indicated no deviations from these assumptions. Collinearity, determined for a model lacking the interactions, appeared to be no issue (maximum Variance Inflation Factor: 1.20) ⁸⁰. We checked for model stability by excluding the levels of the grouping factor plot ID one at a time from the data and comparing the estimates derived with those obtained for the model based on all data ⁸¹, which indicated no influential cases to exist.

We conducted post-hoc pairwise comparisons to compare growth responses to ash dieback between high and low distances to ground water (i.e. on dry and moist sites, respectively) for each species as well as between all species for dry and moist sites, respectively, separately for the lower and upper canopy. To do so, we used the 1000 fitted values obtained through parametric bootstrapping to determine the slopes against ash dieback for the 25% and 75% quantiles of distance to ground water, respectively, and calculated the difference between the slopes per bootstrap. We then determined the proportion of slope differences smaller than or equal to zero and the proportion of slope differences larger than or equal to zero, took the smaller of the two proportions and multiplied it by two to obtain two-tailed p-values.

We fitted the models in R (version 4.3.1) ⁸² using the function lmer of the package lme4 (version 1.1.34) ⁸³ and lmerTest (version 3.1.3) ⁷⁷. We determined Variance Inflation Factors using the function vif of the package car (version 3.1.2) ⁸⁴. We assessed model stability using a function written by RM. We derived confidence intervals using the function bootMer of package lme4, using 1,000 parametric bootstraps and bootstrapping over the random effects, too. We calculated marginal R²-values using the function r.squaredGLMM of the package MuMIn (version 1.47.5) ⁸⁵. We set significance at p < 0.05 and trends at 0.05 ≤ p < 0.1. Computations for this work were done in part using resources of the iDiv High-Performance Computing (HPC) cluster in Leipzig.

The sample size for the LC model encompassed 3585 trees, with 3399 tree IDs nested in 59 plot IDs. The total number of estimated effects for this model was 86, with 41.7 data points per estimated effect. The sample size for the UC model encompassed 851 trees, with 843 tree IDs nested in 59 plot IDs. The total number of estimated effects for this model was 72, with 11.8 data points per estimated effect. Marginal R²-values were 0.196 for both the UC model and the LC model.

3.1. Descriptive statistics

The dominant species in the study area, both in the first and the second inventory of the tree layer (first inventory: N = 7139, second inventory: N = 7610), with decreasing relative dominance, were common ash (Fraxinus excelsior L.), pedunculate oak (Quercus robur L.), sycamore maple (Acer pseudoplatanus L.), small-leaved lime (Tilia cordata Mill.), hornbeam (Carpinus betulus L.), elm (Ulmus spec., including the three elm species field, flutter and mountain elm - Ulmus minor Mill., U. laevis Pall., U. glabra Huds. – and their hybrids, e.g. U. x hollandica Mill.), field maple (Acer campestre L.), and Norway maple (Acer platanoides L.; see Fig. S1 Supplementary Material). F. excelsior, comprising about 40% of the canopy, is the by far most dominant species.

When taking the different tree layers into account, it became evident that the number of individuals and the species composition in the upper canopy were relatively stable over the course of the study period (Figs. 1b and 2a). In the lower canopy, however, all species increased in numbers and relative dominance from the first to the second inventory except F. excelsior and Q. robur whose numbers and relative dominance drastically decreased (Figs. 1a and 2b). In the lower canopy, Acer species together made up 29.8% of the species composition in the first inventory and 38.9% in the second inventory. For presenting descriptive data characterizing stand structure dynamics we use diameter classes. Distributions of annual basal area increments per ha per DBH class showed that while increments in F. excelsior and Q. robur were primarily due to an increase in basal area in high diameter classes (from DBH = 40 cm), increments in the three Acer species – particularly A. pseudoplatanus – were basically based in the advanced regeneration (DBH = 5 to 40 cm), and were 3.5 times higher than those of F. excelsior and Q. robur combined (0.11 vs. 0.03 m² ha^− 1 yr^− 1), already accounting for 24% of the total growth (Fig. S5 Supplementary Material). Recruitment, here defined as the number of individuals (or gain in basal area, respectively) that grew into the DBH class > 5cm, was highest for Ulmus and A. pseudoplatanus and lowest for F. excelsior, while there was no recruitment at all for Q. robur (Fig. S6 Supplementary Material). Median annual relative growth rates (RGR) were highest for A. platanoides, followed by A. campestre, Ulmus spec. and A. pseudoplatanus, and lowest for Q. robur (see Fig. 3 and Table 2). RGRs were generally higher in the lower as compared to the upper canopy for all species except Q. robur (Fig. 3). Overall, distance to ground water was not significantly correlated with ash dieback (Pearson correlation: r= -0.192, t= -1.476, df = 57, p = 0.146, Fig. S8c Supplementary Material).

Table 2

Median and interquartile range of annual relative growth rate (RGR) per species. N = 4726.
species	Median RGR	0.25% quantile	0.75% quantile
A. campestre	0.056	0.026	0.104
A. platanoides	0.076	0.044	0.124
A. pseudoplatanus	0.042	0.019	0.083
C. betulus	0.036	0.018	0.062
Q. robur	0.013	0.007	0.022
T. cordata	0.020	0.008	0.047
Ulmus spec.	0.050	0.023	0.089

3.2. Upper canopy model

Overall, the set of predictor variables had a significant effect on the relative growth rate for trees in the upper canopy (UC; full null model comparison: χ² = 48.200, df = 19, p < 0.001). However, the 3-way interaction between species, ash dieback and distance to ground water was not significant for the UC (F_[4,55.11] = 0.194, p = 0.940, Fig. 4, table S1 Supplementary Material), suggesting no species-specific growth reactions to ash dieback in dependence on hydrological conditions. In order to obtain interpretable p-values for the lower order interactions, a reduced UC model not comprising the 3-way interaction was fitted. The reduced model revealed a significant effect of the interaction between ash dieback and distance to ground water (F_[1,27.96] = 7.929, p = 0.009) but not for the interactions between species and ash dieback (F_[4,40.76] = 0.397, p = 0.810) nor for the interaction between species and distance to ground water (F_[4,38.40] = 1.217, p = 0.320, table S2 Supplementary Material). In order to obtain an interpretable p-value for species, a second reduced model not comprising the non-significant interactions was fitted. This final model revealed a significant effect of species (F_[4,31.78] = 13.095, p < 0.001) and of the interaction between ash dieback and distance to ground water (F_[1,37.66] = 10.497, p = 0.003, table S3 Supplementary Material), suggesting that species generally differed in relative growth rates and that growth responses to ash dieback generally varied depending on hydrological conditions.

As expected from the non-significant three-way interaction between species, ash dieback and distance to ground water, post-hoc pairwise comparisons of growth responses to ash dieback between moist and dry sites and between all species on moist and dry sites did not reveal significance (table S7 and table S8 Supplementary Material).

Furthermore, plot density (estimate=-0.086, F_[1,33.07] = 7.727, p = 0.009, Fig. S9a, table S3 Supplementary Material) and Δdamage degree (estimate=-0.071, F_[1,61.83] = 15.547, p < 0.001, Fig. S9b, table S3 Supplementary Material) had a clear negative effect on relative growth rates in the UC, indicating that growth responses were lower the higher the initial stand density was and the more severely trees got damaged over the years. In contrast, growth responses increased with higher tree vitality in the UC (estimate_{[vitality.normal]} = 0.179, F_[2,32.92] = 4.526, p = 0.018, Fig. S10a, table S3 Supplementary Material). The time interval between inventories did not have a significant effect in the UC (estimate=-0.004, F_[1,42.10] = 0.022, p = 0.882, table S3 Supplementary Material).

3.3. Lower canopy model

Overall, the set of predictor variables had a significant effect on the relative growth rate for trees in the lower canopy (LC; full null model comparison: χ² = 99.175, df = 23, p < 0.001). More specifically, the 3-way interaction between species, ash dieback and distance to ground water was significant for the LC (F_[5,108.75] = 3.327, p = 0.008, Fig. 5, table S4 Supplementary Material), suggesting species-specific growth responses to canopy disturbance depending on hydrological conditions.

Post-hoc pairwise comparisons revealed significant or highly significant differences, respectively, of growth responses to ash dieback between high and low distances to ground water for A. campestre, A. pseudoplatanus and C. betulus in the LC, with all species reacting negatively on dry sites and C. betulus showing the highest differential response (A. campestre: p = 0.034, A. pseudoplatanus: p = 0.006, C. betulus: p < 0.001; table S7 Supplementary Material). When comparing species’ growth responses to ash dieback on moist sites, it became evident that T. cordata reacted significantly different to all other species, with growth responses decreasing with increasing ash dieback, whereas all other species reacted positively to increasing ash dieback on moist sites (see table S9 Supplementary Material, Fig. 5). When comparing the growth responses of all species on dry sites, it became clear that the negative growth response to ash dieback in C. betulus was significantly stronger as compared to most other species (see table S9 Supplementary Material, Fig. 5).

It has to be taken into account, though, that the effect of distance to ground water was based on a continuous variable and comparisons between high and low distances to ground water are conditional on 25% or 75% percentiles for distance to ground water, respectively; thus, effects would be more or less pronounced when higher or lower percentiles were chosen. However, 25% and 75% represent a good amount of data which is why we chose these thresholds.

Similar to the UC, plot density (estimate=-0.078, F_[1,55.60] = 7.347, p = 0.009, Fig. S11a, table S4 Supplementary Material) and Δdamage degree (estimate=-0.083, F_{[1, 55.29]} = 28.545, p < 0.001, Fig. S11b, table S4 Supplementary Material) had a clear negative effect on relative growth rates in the LC, indicating lower growth responses with higher initial stand density and more severe tree damages over the years. In contrast, growth responses increased with higher tree vitality in the LC (estimate_{[vitality.normal]} = 0.331, F(2,67.38) = 50.593, p < 0.001, Fig. S10b, table S4 Supplementary Material). The time interval between inventories had a significantly negative effect in the LC (estimate=-0.069, F_[1,56.34] = 5.100, p = 0.028, table S4 Supplementary Material).

In this study, we aimed to understand growth responses as an important component of the reorganization processes caused by a novel biotic disturbance in combination with novel environmental conditions in a Central European floodplain forest system. Our goal was to determine how the main tree species respond to increasing canopy disturbance caused by ash dieback under different hydrological conditions, while disentangling growth reactions for the upper and lower canopy, respectively. We found seemingly different growth patterns for the lower and the upper canopy that were dependent on hydrological conditions. Our results show that species can only profit from increased light availability under moist hydrological conditions and suffer under dry conditions, and that this effect was species specific and pronounced in the lower canopy but not in the upper canopy. While, in the lower canopy, some species such as the competitive A. pseudoplatanus and A. platanoides profit from ash dieback on moist sites, others are less affected or suffer disproportionally, suggesting that species benefitting from ash dieback now will be an even stronger competitive species to others in the future. Our results give hints on the tree species distribution of the floodplain forest of the future and have important implications for conservation measures, suggesting that revitalization of natural hydrological dynamics is important to maintain a tree composition that resembles the existing one.

4.1. General growth responses to ash dieback and effect of hydrological conditions

Contradicting our first hypothesis, species did not generally profit from increased light availability caused by ash dieback. Whereas growth responses to increasing ash dieback intensity were mostly positive (except for A. campestre and T. cordata in the lower canopy) on moist sites, growth responses to increasing ash dieback intensity on dry sites were mostly negative (except for Ulmus spec. in the lower canopy). This suggests that water availability is a critical factor for tree species to be able to profit from increased light availability due to canopy disturbances and confirms our second hypothesis that growth responses to ash dieback are dependent on hydrological conditions. Most floodplain forest tree species usually develop a shallow root system as an adaptation to the ample water availability in floodplains, relying more on surface water and soil moisture ^86,87, with the disadvantage that especially in dry periods there is limited connectivity to the ground water. Thus, when the groundwater level drops, roots may lose contact to the groundwater and the negative effect of hot summers as experienced during the 2018–2020 consecutive droughts may increase and cause drought stress, which usually results in reduced growth rates ^23,24,88,89. Under drought conditions, trees may allocate resources differently, e.g. prioritizing root growth over radial growth to enhance their ability to access water ⁹⁰. Thus, although canopy disturbance may improve light availability, the remaining trees may only be able to utilize this benefit after a significant delay – if at all – because root expansion is necessary first to alleviate the constraints on water supply ⁹¹. As the time interval between our two inventories was quite short, not enough time might have passed for the trees to adapt to the new conditions but it is also possible that under prolonged drought trees might not benefit at all from increased light availability but rather suffer due to increased drought stress. In the same forest, Schnabel et al. (2022) ²⁴ found strong decreases in tree growth and increases in physiological stress responses to consecutive droughts, and there is ample evidence for negative effects of droughts in other temperate floodplain forests on tree growth, inducing increased stress and higher tree mortality ^23,88,89. Thus, it is likely that the compound event of ash dieback and climate extremes might cause the negative growth reactions observed on dry sites in our study.

Where the distance to groundwater is low, i.e. with good water supply, however, increased light availability and warmer temperatures can be used to increase the rate of photosynthesis, as the transpiration losses can be quickly compensated for, which might explain the positive growth responses to ash dieback we find on moist sites. Multiple studies found evidence for positive growth responses of residual trees to canopy disturbance ^26,35,39,45, and access to groundwater was found to buffer negative effects of summer drought on floodplain forest tree growth, at least in oak trees ²⁵. Furthermore, water from the groundwater is available for fine roots through capillary rise when the distance to ground water is low ⁹², and unpublished data from our forest ecosystem suggest that soil moisture is strongly correlated with distance to groundwater level (Vieweg et al. unpublished data), making distance to ground water a good predictor for soil moisture. However, positive growth reactions on moist sites were generally quite weak which might again be due to the relatively short time interval between the two inventories so that not enough time has passed yet to physiologically adapt to the new environmental conditions ^34,36. Alternatively, it is possible that trees may not be capable of utilizing improved resource availability because they are suffering from drought legacy effects of the 2018 and 2019 consecutive drought years ^24,93. It is important to note that our results represent the effects of a compound event of a pathogen invasion and climate extremes opening up the canopy via ash mortality, and the different behaviors under moist and dry conditions could in fact be interpreted as being approximately representative of past and future climatic conditions, respectively. The effects we see on dry sites can actually be an indicator of reorganization processes under progressing climate change and canopy disintegration. It further has to be noted that, during our study period, the ash dieback was just at its beginnings and accelerated tremendously after 2020 ⁶³. Thus, we expect effects to be much stronger if another inventory is carried out now.

Our result that growth responses significantly depend on hydrological conditions is in line with our hypothesis but contrasts previous findings in the Leipzig floodplain forest which only showed a minor effect of hydrological conditions on tree growth during the 2018–2019 consecutive drought ²⁴. This might be due to the fact that for our study, we had more fine-scaled tree-based data on distance to ground water based on a new coupled hydrological model ⁵⁶ and a higher sample size available, whereas Schnabel et al. (2022) ²⁴ relied on a broader categorization of hydrological strata based on the plot level which was due to limitations in the initial phase of the “Lebendige Luppe” project, as well as on a smaller sample size. Furthermore, Schnabel et al. (2022) ²⁴ focused on a shorter study period only encompassing extreme drought years; thus, soil water availability was likely generally low due to the consecutive droughts in the recent years and the hydrological gradient insufficient to explain variation in tree growth.

4.2. Growth responses in the lower and upper canopy

We expected particularly strong positive effects of ash dieback on growth responses in the lower canopy, as the increase in light availability is relatively higher in the lower as compared to the upper canopy. At the same time, we expected weak responses to ash dieback for trees in the upper canopy as they are less likely to be limited by competition for light. Our results show seemingly different growth patterns for the upper and lower canopy. In the lower canopy, the significant three-way interaction revealed a strong species-specific dependency of growth responses on both ash dieback intensity and hydrological conditions. However, contrary to our third hypothesis, we did not find particularly pronounced positive growth reactions but rather pronounced differences between moist and dry sites, which were particularly strong in A. campestre, A. pseudoplatanus and C. betulus, and which may be representative for the high differences between “past or normal” and “future or novel” climatic conditions. In the upper canopy, in line with our fourth hypothesis, the effect of ash dieback and hydrological conditions was less pronounced as the three-way interaction revealed no species-specific reaction to ash dieback depending on hydrological conditions and post-hoc comparisons showed no significant differences between moist and dry sites for any of the species. This is probably related to the larger rooting depth of overstorey trees that can buffer hydrological variations better as compared to the roots of smaller trees of the lower canopy. Our results suggest that growth responses to different hydrological conditions and ash dieback are not species specific in the upper canopy, whereas species respond differentially to environmental changes in the lower canopy. Increasing canopy disturbance can lead to the loss of the canopy’s buffering function against climate extremes ^48,94–96 and cause greater microclimatic changes in the lower canopy as compared to the upper canopy. For lower canopy trees, the increasing exposure leads to an increasing coupling with the atmosphere, resulting in increased radiation and air temperature, possibly reducing air moisture and increasing vapor pressure deficit (VPD), which in turn might lead to faster stomatal closure and decreasing photosynthesis and growth rates and eventually to mortality due to carbon starvation ^97–100. Larger trees in the upper canopy may furthermore be less capable of responding to the increased resource availability in dependence on hydrology due to physiological constraints associated with age, as they are already longer adapted to their site conditions than smaller trees ^34,41.

These negative effects on growth might even deteriorate as soil water availability decreases, as supported by our findings. The pronounced negative effects of ash dieback on growth responses for the lower canopy on dry sites are in line with recent findings by Meyer et al. (2022) ¹⁰¹, who found that higher mortality in unmanaged stands during the 2018 and 2019 droughts primarily affected suppressed understory trees and is likely a result of the combination of competition and drought-related stress. Mortality is usually preceded by growth declines indicative of inciting stress that can span several decades ^102–104. Smaller trees in the lower canopy may furthermore have limited access to deeper water layers due to their less developed root systems ¹⁰⁵ and may thus be more prone to drought-induced stress. Furthermore, in the lower canopy, some species such as C. betulus and T. cordata suffer disproportionally from the novel dry conditions as compared to other species, above all Acer. Our results suggest that if soil conditions continue to dry up with proceeding climate change, growth decline might ultimately lead to increased mortality for some species but not for others in the lower canopy and thus change forest composition, structure and functioning.

Previous studies showed that smaller overstory trees which are growing in suppressed conditions created through shading by larger overstory trees show a relatively greater growth response to improved light access than larger trees, especially when they grow near a canopy gap ^35,37,38 (but see ^42,106). We did not find a particularly pronounced positive growth reaction to increased light availability in the lower canopy. It has to be noted, though, that as we could not integrate canopy layer as an interaction effect in our model due to an unbalanced representation of several species in both tree layers, we cannot directly test for different growth responses between the lower and upper canopy. However, the distinct patterns we found in the different tree layers suggest that different underlying mechanisms are at play. What became evident, though, was that growth responses were in general higher in the lower as compared to the upper canopy, which is in line with previous findings showing that growth rates decline with increasing tree age and size ^{107, but see 108}.

4.3. Species specific growth responses in the light of shade and drought tolerance

We expected growth responses to ash dieback and hydrological conditions to be species specific. Consistent with this hypothesis, we found species-specific growth responses to ash dieback and hydrological conditions in the lower canopy but not in the upper canopy where species in general showed differential growth responses but not in relation to ash dieback and distance to groundwater. Species-specific responses to varying environmental conditions can be explained by differences in shade or drought-tolerance and specific water demands. Although, according to Ellenberg’s light indicator values ¹⁰⁹, our examined tree species show similar levels of shade and drought tolerance, being intermediate for all species except for Q. robur which stands out due to its low shade tolerance and high drought tolerance, there is some variation, with A. campestre and T. cordata being slightly more light demanding (L5) than the remaining species (L4) (see table S10 Supplementary Material). A. campestre and T. cordata, the less shade-tolerant species, profited the least from canopy disturbance on moist sites in the lower canopy. T. cordata significantly differed from all other species in its reaction to ash dieback on moist sites, being the only species showing pronounced negative growth reactions when water availability was sufficient. Our results are in line with previous findings showing that more shade-tolerant species respond more positively to increased light availability due to canopy gaps than less shade-tolerant species, especially smaller individuals in the lower canopy ³⁵ (but see ⁴⁵). In general, shade-tolerant species are more likely to respond to small gap openings as their physiological and morphological plasticity enables them to quickly adjust to increased light availability ³², which might explain the more positive growth responses of more shade-tolerant species on moist sites in our study. However, it has to be acknowledged that the pattern we found on moist sites is subtle and that the distinct negative reactions of T. cordata could also be attributed to high density competition by Acer species, especially A. pseudoplatanus.

On dry sites, C. betulus showed the most pronounced negative growth response to ash dieback intensity, which differed significantly from growth responses of all other species except T. cordata. This finding is somewhat surprising, as C. betulus is thought to have a moderate to high level of drought tolerance in the adult stage ¹¹⁰ (see table S10 Supplementary Material) and high drought resistance in the juvenile stage ¹¹¹ that is not below that of the other species. The negative reactions might most likely be the result of indirect competitive interactions between the species, as are the pronounced negative responses of T. cordata on both moist and dry sites. Although A. pseudoplatanus did not exhibit increased growth responses that were significantly higher than those of other species, demographic data from our two inventories show that A. pseudoplatanus is highly abundant, both in the regeneration layer and the lower canopy, that its abundance increased over the study period and that it shows very high recruitment ^21,112,113 (Figs. 1a and 2b, fig. S6 Supplementary Material). Although we cannot confirm this based on the results of our model, it is well conceivable that A. pseudoplatanus might outcompete other species very quickly not because it grows more rapidly than other species in the lower canopy but simply because of a fast recruitment and thus high abundance. The high growth increments of A. pseudoplatanus observed in advanced regeneration (in the lower DBH classes from 5 to 40 cm, see Fig. S5 Supplementary Material) indicate a shift in forest development towards this competitive species that could result in progressing competitive take over, especially if hydrological conditions worsen in the light of climate change, favouring the flood-intolerant A. pseudoplatanus even more. Ulmus seemed to be unaffected by ash dieback in the lower canopy, both under moist and dry hydrological conditions, which is likely due to its vegetative reproduction mode that makes them less sensitive to changing and adverse environmental conditions ¹¹⁴. Elm trees are basically non-existent anymore in the upper canopy due to the Dutch elm disease which increasingly spread in the 1960s, but are still abundant in the regeneration layer and lower canopy ^21,112 (see Fig. 1). Although most Ulmus species will most likely not play a role in the upper canopy anymore as trees die when reaching the adult phase due to the fungal disease, our findings suggest that Ulmus is likely to persist in the lower canopy even under adverse novel environmental circumstances.

It would have been interesting to see how Q. robur as the most light-demanding species (L7) in our sample reacts to increased light availability in the lower canopy. However, Q. robur is hardly present in the lower canopy and almost absent in the regeneration layer ^{21,112,115,116} (see Fig. 1c&d), as it is no longer able to successfully regenerate under natural circumstances, mostly due to the limited availability of light on the forest floor which is exacerbated by the dominating spread of shadecasting Acer species ^30,52,70 (see Fig. S6 Supplementary Material). We could therefore not reliably model its growth responses to increased light availability in the lower canopy. In the upper canopy, Q. robur did not seem to perform any worse or better to increased light availability compared to the other species, although growth rates seemed to be generally lower in Q. robur than in other species (see Fig. 2, Table 2). From an ecological and conservation perspective, Q. robur is of utmost importance as it supports a very high and specific biodiversity for a wide range of species groups ^117–120 and is particularly vital for many threatened species ^121,122. Thus, its preservation is necessary for the ecosystem’s conservation status.

4.4. Implications for conservation

Floodplain forests worldwide show an unfavourable to bad conservation status ^30,123, threatening their role as biodiversity hotspots ^13,15 and the numerous ecosystem functions they provide ^{16,17,124,125}, and are thus highly protected on a global and local scale ^13,52. Even small changes in the water dynamics can lead to a drastically altered species composition and diversity ¹³. Our results suggest that groundwater levels are a crucial factor influencing growth reactions of floodplain forest tree species to canopy disturbance caused by ash dieback. They further confirm the importance of groundwater as a proxy for floodplain dynamics and as essential abiotic driver for floodplain tree performance. Low distances to groundwater appear to mitigate negative effects of canopy disturbances on residual tree growth, highlighting the importance of revitalizing hydrodynamics and raising groundwater levels in floodplain forest ecosystems that dried up due to anthropogenic influence, especially in the light of ongoing climate change and more frequent occurrences of extreme drought events in the future. According to our findings, Acer species, and here especially A. pseudoplatanus and A. platanoides, are likely to dominate the future species composition in floodplain forests, as they exhibit high growth rates that do not decline as much under dry and light conditions as compared to those of other species such as C. betulus or T. cordata. Our results also suggest that if there will only be a moderate degree of rewetting without a substantial revitalization of the floodplain, this may inadvertently favor A. pseudoplatanus instead of displacing it. The duration of rewetting must be sufficient to selectively increase mortality substantially for flood-intolerant Acer species, thus removing Acer from the system and release other species from density competition by Acer.

Access to ground water was found to buffer the negative effects of increasing summer water-deficits on floodplain forest tree growth before ²⁵. If revitalization measures are not being implemented soon enough and the extent of tree diseases and mortality further increases, in combination with sinking groundwater levels due to climate change, it is likely that floodplain forests will experience increasing stress and might turn into a novel ecosystem dominated by competitive Acer species that are untypical for hardwood floodplain forests. With the loss of Ulmus and now F. excelsior and the continuous decline of Q. robur within the past decades, the ecosystem might lose its three characteristic main tree species and with them countless threatened species that depend on them. As a consequence, the conservation status of hardwood floodplain forests as Natura 2000 sites is threatened.

Alternatively, if revitalization measures, such as raising riverbeds, reactivating and integrating antiquated watercourses, dismantling or perforating dikes ³¹, will be implemented soon enough and restore natural flooding dynamics and elevated groundwater levels, together with nature conservation-compliant promotion of oak regeneration, the hardwood forest community might have a chance to maintain a state that at least resembles the characteristic composition. Although we found that the intensity of ash dieback does not seem to depend on hydrological conditions (see Fig. S8c Supplementary Material) and thus revitalization might not stop or ameliorate the disease, resistant genotypes or epigenetic responses might keep ash trees in the system in the medium term ^126–128. Our results indicate that Acer growth responses to canopy disturbance are not as exceedingly high as expected based on the demographic data and might curtail its dominance together with an increased mortality due to the sooty bark disease (Cryptostroma corticale) ^21,70. A reestablished natural flooding regime might furthermore suppress Acer regeneration. Areas of increased mortality and thus increased light availability could be used to actively promote the light-demanding Q. robur through planting and removal of Acer species, as it was shown that the vitality of Q. robur saplings is highest under these conditions (Lenk et al. unpublished data).

Our study looked at growth as one component of forest reorganization processes and focused on early responses after the canopy has just began to open up due to ash dieback. It is crucial to scientifically monitor these immediate effects from the beginning as they give important insights on early growth responses of different tree species unaffected by competitive interactions among them. However, it is important to note that reorganization not only depends on growth, but also on other demographic rates such as mortality and recruitment that we did not consider here. In order to fully understand the mechanisms of forest reorganization processes, it is of utmost importance to continue the observations and to include all demographic rates. Currently, efforts are made to model these dynamics in more detail using field-parameterized demographic growth models ¹²⁹. Furthermore, as the ash dieback was just at its beginnings during our study period and accelerated tremendously after 2020, we would expect effects to be much stronger should another inventory be carried out now.

The results of our scenarios of moist and dry site conditions in our study could also be interpreted as being representative of “past/ normal” and “future/ novel” hydrological states. The negative effects we found on dry sites might even underestimate what happens under progressing climate change and more frequent and intense droughts, as trees on moist sites or under “normal” conditions, respectively, will potentially suffer even more when faced with a sudden changing hydrology due to droughts as compared to the trees on our dry sites that already adapted to these dry conditions for decades. Thus, our results most likely represent a rather conservative estimation. If abiotic conditions change substantially in the future, it is also possible that other previously absent or underrepresented species, such as the European beech, might establish in the system and drastically alter forest composition and functioning. Findings from this study and related research could help predict how tree species and forests react to these new climatic conditions.

Competing interests

The authors declare no competing interests.

Additional information

Supplementary Material

Author Contribution

SH, CW, CSD, RR and KA conceptualized the study design. CSD, CW, MS, HK designed the plot network. MS initialized the ash dieback inventory. GR, CSD, RE, TH, CH, MV and MD collected the data. SH cleaned the data with contributions from GR. SH, RM, RR and KA analyzed the data. SH wrote the manuscript. All authors discussed the results and revised the manuscript.

Acknowledgement

We would like to thank our colleagues of the ‘Lebendige Luppe’ (living Luppe river) project, coordinated by Angela Zábojník, for a successful collaboration and establishment of the plot network. We thank Andreas Sickert and Martin Opitz (Stadtforst) as well as Andreas Padberg and Carsten Pitsch (Sachsenforst) for their continuous support throughout the project. We further thank all field assistants for contributing to the data collection as well as Christian Krause from the German Centre for Integrative Biodiversity Research (iDiv) Halle-Jena-Leipzig for assistance in using the High-Performance Computing (HPC) cluster for data analysis. The Lebendige Luppe project was funded by the German Federal Agency for Nature Conservation (BfN) in cooperation with the German Federal Ministry for the Environment, Nature Conservation and Nuclear Safety (BMU) as part of the Federal Biological Diversity Programme. The state of Saxony supported the project through the nature conservation fund of the Saxon State Foundation for Nature and the Environment (LaNU). Further information can be found on the project homepage www.Lebendige-Luppe.de.

Data Availability

Tree inventory data are currently under review at Pangaea. The DOI is https://doi.pangaea.de/10.1594/. Ash dieback inventory data are in the process of submission to Pangaea and will be openly available soon. The DOI follows.R-scripts and functions are available upon reasonable request. Please contact Stefanie Henkel at [email protected].

Seidl, R. & Turner, M. G. Post-disturbance reorganization of forest ecosystems in a changing world. Proc. Natl. Acad. Sci. 119, e2202190119 (2022).
Senf, C. & Seidl, R. Post‐disturbance canopy recovery and the resilience of Europe’s forests. Glob. Ecol. Biogeogr. 31, 25–36 (2022).
Allen, C. D. et al. A global overview of drought and heat-induced tree mortality reveals emerging climate change risks for forests. For. Ecol. Manag. 259, 660–684 (2010).
Senf, C., Sebald, J. & Seidl, R. Increasing canopy mortality affects the future demographic structure of Europe’s forests. One Earth 4, 749–755 (2021).
Turner, M. G. Disturbance and landscape dynamics in a changing world. Ecology 91, 2833–2849 (2010).
Batllori, E. et al. Forest and woodland replacement patterns following drought-related mortality. Proc. Natl. Acad. Sci. 117, 29720–29729 (2020).
Rammer, W. et al. Widespread regeneration failure in forests of Greater Yellowstone under scenarios of future climate and fire. Glob. Change Biol. 27, 4339–4351 (2021).
Millar, C. I. & Stephenson, N. L. Temperate forest health in an era of emerging megadisturbance. Science 349, 823–826 (2015).
Bravard, J.-P., Amoros, C. & Pautou, G. Impact of Civil Engineering Works on the Successions of Communities in a Fluvial System: A Methodological and Predictive Approach Applied to a Section of the Upper Rhône River, France. Oikos 47, 92 (1986).
Haase, D. & Gläser, J. Determinants of floodplain forest development illustrated by the example of the floodplain forest in the District of Leipzig. For. Ecol. Manag. 258, 887–894 (2009).
Kowalska, N. et al. Analysis of floodplain forest sensitivity to drought. Philos. Trans. R. Soc. B Biol. Sci. 375, 20190518 (2020).
Leuschner, C. & Ellenberg, H. Ecology of Central European Forests: Vegetation Ecology of Central Europe, Volume I. (Springer International Publishing, Cham, 2017). doi:10.1007/978-3-319-43042-3.
Tockner, K. & Stanford, J. A. Riverine flood plains: present state and future trends. Environ. Conserv. 29, 308–330 (2002).
BMU & BfN. Auenzustandsbericht-Flussauen in Deutschland. (2021).
Ellenberg, H. & Leuschner, C. Vegetation Mitteleuropas mit den Alpen: in ökologischer, dynamischer und historischer Sicht. (UTB, Stuttgart, Stuttgart, 2010).
Haase, D. Holocene floodplains and their distribution in urban areas—functionality indicators for their retention potentials. Landsc. Urban Plan. 66, 5–18 (2003).
Havrdová, A., Douda, J. & Doudová, J. Threats, biodiversity drivers and restoration in temperate floodplain forests related to spatial scales. Sci. Total Environ. 854, 158743 (2023).
Janik, D. et al. Tree layer dynamics of the Cahnov–Soutok near-natural floodplain forest after 33 years (1973–2006). Eur. J. For. Res. 127, 337–345 (2008).
Pörtner, H.-O. et al. IPCC, 2022: Summary for Policymakers [H.-O. Pörtner, D.C. Roberts, E.S. Poloczanska, K. Mintenbeck, M. Tignor, A. Alegría, M. Craig, S. Langsdorf, S. Löschke, V. Möller, A. Okem (eds.)]. In: Climate Change 2022: Impacts, Adaptation and Vulnerability. Contribution of Working Group II to the Sixth Assessment Report of the Intergovernmental Panel on Climate Change [H.-O. Pörtner, D.C. Roberts, M. Tignor, E.S. Poloczanska, K. Mintenbeck, A. Alegría, M. Craig, S. Langsdorf, S. Löschke, V. Möller, A. Okem, B. Rama (eds.)]. in Climate Change 2022: Impacts, Adaptation and Vulnerability. Contribution of Working Group II to the Sixth Assessment Report of the Intergovernmental Panel on Climate Change (eds. Pörtner, H.-O. et al.) 3–33 (Cambridge University Press, Cambridge, UK and New York, NY, USA, 2022).
Rakovec, O. et al. The 2018–2020 Multi-Year Drought Sets a New Benchmark in Europe. Earths Future 10, e2021EF002394 (2022).
Wirth, C. et al. Ein Biodiversitätshotspot an der Belastungsgrenze - Naturschutz und Klimawandel im Leipziger Auwald. Biol. Unserer Zeit - BiuZ 55–65 (2021) doi:10.11576/biuz- 4107.
Hari, V., Rakovec, O., Markonis, Y., Hanel, M. & Kumar, R. Increased future occurrences of the exceptional 2018–2019 Central European drought under global warming. Sci. Rep. 10, 12207 (2020).
Colangelo, M. et al. Drought Decreases Growth and Increases Mortality of Coexisting Native and Introduced Tree Species in a Temperate Floodplain Forest. Forests 9, 205 (2018).
Schnabel, F. et al. Cumulative growth and stress responses to the 2018–2019 drought in a European floodplain forest. Glob. Change Biol. 28, 1870–1883 (2022).
Skiadaresis, G., Schwarz, J. A. & Bauhus, J. Groundwater Extraction in Floodplain Forests Reduces Radial Growth and Increases Summer Drought Sensitivity of Pedunculate Oak Trees (Quercus robur L.). Front. For. Glob. Change 2, 5 (2019).
Flower, C. E. & Gonzalez-Meler, M. A. Responses of Temperate Forest Productivity to Insect and Pathogen Disturbances. Annu. Rev. Plant Biol. 66, 547–569 (2015).
Weed, A. S., Ayres, M. P. & Hicke, J. A. Consequences of climate change for biotic disturbances in North American forests. Ecol. Monogr. 83, 441–470 (2013).
Klimo & Hager. The Floodplain Forests in Europe: Current Situation and Perspectives. (Koninklijke Brill NV; Leiden, The Netherlands, Brill: Leiden, Boston, Köln, 2001., 2001).
Broome, A., Ray, D., Mitchell, R. & Harmer, R. Responding to ash dieback ( Hymenoscyphus fraxineus ) in the UK: woodland composition and replacement tree species. For. Int. J. For. Res. 92, 108–119 (2019).
Härdtle, W. et al. Pflanzengesellschaft des Jahres 2021: Hartholz-Auenwald (Ficario-Ulmetum). (2020) doi:10.14471/2020.40.007.
Wirth, C. et al. Dynamik als Leitprinzip zur Revitalisierung des Leipziger Auensystems: 10 Thesen zur Revitalisierung der Leipziger Aue, eine Vision, ein konkreter Maßnahmenkatalog mit Karte zu Dynamisierungsoptionen und ein Ausblick mit Realisierungsvorschlägen. Dtsch. Zent. Für Integr. Biodiversitätsforschung IDiv UFZ-Dep. Naturschutzforschung 64 (2020).
McCarthy, J. Gap dynamics of forest trees: A review with particular attention to boreal forests. Environ. Rev. 9, 1–59 (2001).
Schliemann, S. A. & Bockheim, J. G. Methods for studying treefall gaps: A review. For. Ecol. Manag. 261, 1143–1151 (2011).
Oliver, C. D. & Larson, B. A. Forest Stand Dynamics, Update Edition. (FES Other Publications., 1996).
Jones, T. A., Domke, G. M. & Thomas, S. C. Canopy tree growth responses following selection harvest in seven species varying in shade tolerance. Can. J. For. Res. 39, 430–440 (2009).
Oliver, C. D. & Murray, M. D. Stand structure, thinning prescriptions, and density indexes in a Douglas-fir thinning study, Western Washington, U.S.A. Can. J. For. Res. 13, 126–136 (1983).
Pedersen, B. S. & Howard, J. L. The influence of canopy gaps on overstory tree and forest growth rates in a mature mixed-age, mixed-species forest. For. Ecol. Manag. 196, 351–366 (2004).
Wiser, S. K., Allen, R. B., Benecke, U., Baker, G. & Peltzer, D. Tree growth and mortality after small-group harvesting in New Zealand old-growth Nothofagus forests. Can. J. For. Res. 35, 2323–2331 (2005).
Costilow, K. C., Knight, K. S. & Flower, C. E. Disturbance severity and canopy position control the radial growth response of maple trees (Acer spp.) in forests of northwest Ohio impacted by emerald ash borer (Agrilus planipennis). Ann. For. Sci. 74, 10 (2017).
Muth, C. C. & Bazzaz, F. A. Tree canopy displacement at forest gap edges. Can. J. For. Res. 32, 247–254 (2002).
Stan, A. B. & Daniels, L. D. Growth releases across a natural canopy gap-forest gradient in old-growth forests. For. Ecol. Manag. 313, 98–103 (2014).
Pretzsch, H. Canopy space filling and tree crown morphology in mixed-species stands compared with monocultures. For. Ecol. Manag. 327, 251–264 (2014).
Brüllhardt, M., Rotach, P., Bigler, C., Nötzli, M. & Bugmann, H. Growth and resource allocation of juvenile European beech and sycamore maple along light availability gradients in uneven-aged forests. For. Ecol. Manag. 474, 118314 (2020).
Forrester, J. A., Lorimer, C. G., Dyer, J. H., Gower, S. T. & Mladenoff, D. J. Response of tree regeneration to experimental gap creation and deer herbivory in north temperate forests. For. Ecol. Manag. 329, 137–147 (2014).
Yoshida, T. & Kamitani, T. Effects of Crown Release on Basal Area Growth Rates of Some Broad-Leaved Tree Species with Different Shade-Tolerance. J. For. Res. 3, 181–184 (1998).
Loewenstein, N. J. & Pallardy, S. G. Drought tolerance, xylem sap abscisic acid and stomatal conductance during soil drying: a comparison of canopy trees of three temperate deciduous angiosperms. Tree Physiol. 18, 431–439 (1998).
Ward, J. V., Tockner, K. & Schiemer, F. Biodiversity of floodplain river ecosystems: ecotones and connectivity1. Regul. Rivers Res. Manag. 15, 125–139 (1999).
Richter, R. et al. Tree species matter for forest microclimate regulation during the drought year 2018: disentangling environmental drivers and biotic drivers. Sci. Rep. 12, 17559 (2022).
Scholz, M. et al. Das Projekt Lebendige Luppe – ein Beitrag zur Renaturierung der Leipziger Nord-West-Aue. Auenmagazin 14–21 (2018).
Kasperidus, H. D. & Scholz, M. Auen und Auenwälder in urbanen Räumen. in Der Leipziger Auwald - ein dynamischer Lebensraum; Tagungsband zum 5. Leipziger Auensymposium am 16. April 2011 (eds. Wirth, C., Reiher, A., Zäumer, U. & Kasperidus, H. D.) 26–30 (Helmholtz-Zentrum für Umweltforschung - UFZ, Leipzig, 2011).
Günther-Diringer, D. et al. Methodische Grundlagen zum Auenzustandsbericht 2021: Erfassung, Bilanzierung und Bewertung von Flussauen. (Deutschland / Bundesamt für Naturschutz, 2021). doi:10.19217/skr591.
Gutte, P. Das Querco-Ulmetum minoris Issler 1942, der Stieleichen- Ulmen-Hartholzwald, in der Elster-Luppe-Aue bei Leipzig. Mauritiana 22, 213–242 (2011).
Scholz, M. et al. Das Projekt Lebendige Luppe. Einführung in Den Projektraum Elster-Luppe-Aue. 7–20 (2022).
Kirsten, F., Herkelrath-Bleyl, A., Krüger, A. & Heinrich, J. Entstehung Und Eigenschaften Der Böden Und Sedimente in Der Elster-Luppe-Aue. 43–68 (2022).
Seele-Dilbat, C. et al. Untersuchungsdesign der naturwissenschaftlichen Begleitung im Projekt Lebendige Luppe. UFZ-Ber. 21–42 (2022).
Brückner, F., Sahlbach, T., Buschmann, T. & Vieweg, M. Gekoppeltes Grundwasser-Oberflächenwasser-Modell. (2024).
Vieweg, M., Brückner, F., Kasperidus, H. D., Scholz, M. & Krieg, R. Einrichtung Und Monitoring Des Grund- Und Oberflächenwassers. (2024).
Gläser, J. & Schmidt, P. A. Zur historischen Entwicklung des Baumartenbestandes von Hartholz-Auenwäldern - dargestellt am Beispiel des Leipziger Auenwaldes. Allg. Forst- Jagdztg. 178, 90–97 (2007).
Kuntze, O. Taschen-Flora von Leipzig. (Leipzig, Heidelberg, 1867).
Rieland, G. et al. Tree inventory dataset of floodplain forest, Leipzig, Germany. PANGAEA (2024).
Seele-Dilbat, C., Pruschitzki, U., Engelmann, R., Scholz, M. & Wirth, C. Anweisung Waldinventur - Projekt „Lebendige Luppe" - Attraktive Auenlandschaft als Leipziger Lebensader – Biologische Vielfalt bringt Lebensqualität in die Stadt. (2020).
Grala-Michalak, J. & Kaźmierczak, K. Discriminant analysis for Kraft’s classes of trees. Biom. Lett. 48, 67–81 (2011).
Scholz, M. et al. Die Entwicklung Des Eschentriebsterbens von 2016 Bis 2023 Im Leipziger Auwald. (2024).
Langer, G. & Bressem, U. Eschentriebsterben. Nordwestdtsch. Forstl. Vers. NW-FVA Abt. Waldschutz 4, 1–30 (2016).
Lenz, H., Straßer, L., Baumann, M. & Baier, U. Boniturschlüssel zur Einstufung der Vitalität von Alteschen. AFZ-DerWald 18–19 (2012).
Peters, S., Langer, G. & Kätzel, R. Bonitur geschädigter Eschen im Kontext des Eschentriebsterbens. AFZ-DerWald 12, 28–31 (2021).
Peters, S., Langer, G. & Kätzel, R. Eschentriebsterben - Kriterien zur Schadensbonitur an Eschen. (Fachagentur Nachwachsende Rohstoffe e. V. (FNR), Gülzow-Prüzen, 2021).
Erfmeier, A. et al. Ash Dieback and Its Impact in Near-Natural Forest Remnants – A Plant Community-Based Inventory. Front. Plant Sci. 10, 658 (2019).
Bartha, B., Lenz, H. & Petercord, R. Keine Entwarnung beim Eschentriebsterben. LWF Aktuell 101, 51–53 (2014).
Engelmann, R. A. et al. Reiner Prozessschutz gefährdet Artenvielfalt im Leipziger Auwald. UFZ Discuss. Pap. (2019).
Baayen, R. H. Analyzing Linguistic Data. A Practical Introduction to Statistics. (Cambridge University Press, Cambridge, 2008).
Barr, D. J., Levy, R., Scheepers, C. & Tily, H. J. Random effects structure for confirmatory hypothesis testing: Keep it maximal. J. Mem. Lang. 68, 10.1016/j.jml.2012.11.001 (2013).
Schielzeth, H. & Forstmeier, W. Conclusions beyond support: overconfident estimates in mixed models. Behav. Ecol. 20, 416–420 (2009).
Forstmeier, W. & Schielzeth, H. Cryptic multiple hypotheses testing in linear models: overestimated effect sizes and the winner’s curse. Behav. Ecol. Sociobiol. 65, 47–55 (2011).
Dobson, A. J. An Introduction to Generalized Linear Models. (Chapman & Hall/CRC, Boca Raton, 2002).
Luke, S. G. Evaluating significance in linear mixed-effects models in R. Behav. Res. Methods 49, 1494–1502 (2017).
Kuznetsova, A., Brockhoff, P. B. & Christensen, R. H. B. lmerTest Package: Tests in Linear Mixed Effects Models. J. Stat. Softw. 82, 1–26 (2017).
Schielzeth, H. Simple means to improve the interpretability of regression coefficients. Methods Ecol. Evol. 1, 103–113 (2010).
Field, A. Discovering Statistics Using SPSS, 2nd Ed. xxxiv, 779 (Sage Publications, Inc, Thousand Oaks, CA, US, 2005).
Quinn, G. P. & Keough, M. J. Experimental Design and Data Analysis for Biologists. (Cambridge University Press, Cambridge, 2002). doi:10.1017/CBO9780511806384.
Nieuwenhuis, R., Grotenhuis, M., te & Pelzer, B. influence.ME: Tools for Detecting Influential Data in Mixed Effects Models. R J. 4, 38 (2012).
R Core Team. R: A language and environment for statistical computing. R Foundation for Statistical Computing (2022).
Bates, D., Mächler, M., Bolker, B. & Walker, S. Fitting Linear Mixed-Effects Models Using lme4. J. Stat. Softw. 67, 1–48 (2015).
Fox, J. & Weisberg, S. An R Companion to Applied Regression. (Sage, Thousand Oaks CA, 2019).
Bartón, K. MuMIn: Multi-Model Inference. (2023).
Sánchez-Pérez, J. M., Lucot, E., Bariac, T. & Trémolières, M. Water uptake by trees in a riparian hardwood forest (Rhine floodplain, France). Hydrol. Process. 22, 366–375 (2008).
Singer, M. B. et al. Floodplain ecohydrology: Climatic, anthropogenic, and local physical controls on partitioning of water sources to riparian trees. Water Resour. Res. 50, 4490–4513 (2014).
Mikac, S. et al. Drought-induced shift in tree response to climate in floodplain forests of Southeastern Europe. Sci. Rep. 8, 16495 (2018).
Stojanović, D. B., Levanič, T., Matović, B. & Orlović, S. Growth decrease and mortality of oak floodplain forests as a response to change of water regime and climate. Eur. J. For. Res. 134, 555–567 (2015).
Brunner, I., Herzog, C., Dawes, M. A., Arend, M. & Sperisen, C. How tree roots respond to drought. Front. Plant Sci. 6, (2015).
Pretzsch, H. Grundlagen der Waldwachstumsforschung. (Springer Berlin Heidelberg, Berlin, Heidelberg, 2019). doi:10.1007/978-3-662-58155-1.
Mu, M. et al. Exploring how groundwater buffers the influence of heatwaves on vegetation function during multi-year droughts. Earth Syst. Dyn. 12, 919–938 (2021).
Schuldt, B. et al. A first assessment of the impact of the extreme 2018 summer drought on Central European forests. Basic Appl. Ecol. 45, 86–103 (2020).
Alkama, R. & Cescatti, A. Biophysical climate impacts of recent changes in global forest cover. Science 351, 600–604 (2016).
Davis, K. T., Dobrowski, S. Z., Holden, Z. A., Higuera, P. E. & Abatzoglou, J. T. Microclimatic buffering in forests of the future: the role of local water balance. Ecography 42, 1–11 (2019).
De Frenne, P. et al. Forest microclimates and climate change: Importance, drivers and future research agenda. Glob. Change Biol. 27, 2279–2297 (2021).
Facciano, L., Sasal, Y. & Suarez, M. L. How do understory trees deal with small canopy openings? The case of release in growth following drought-induced tree mortality. For. Ecol. Manag. 529, 120692 (2023).
Grossiord, C. et al. Plant responses to rising vapor pressure deficit. New Phytol. 226, 1550–1566 (2020).
McDowell, N. G. Mechanisms Linking Drought, Hydraulics, Carbon Metabolism, and Vegetation Mortality. Plant Physiol. 155, 1051–1059 (2011).
Rollinson, C. R. et al. Climate sensitivity of understory trees differs from overstory trees in temperate mesic forests. Ecology 102, e03264 (2021).
Meyer, P., Spînu, A. P., Mölder, A. & Bauhus, J. Management alters drought-induced mortality patterns in European beech (Fagus sylvatica L.) forests. Plant Biol. 24, 1157–1170 (2022).
Cailleret, M. et al. A synthesis of radial growth patterns preceding tree mortality. Glob. Change Biol. 23, 1675–1690 (2017).
Gillner, S., Rüger, N., Roloff, A. & Berger, U. Low relative growth rates predict future mortality of common beech (Fagus sylvatica L.). For. Ecol. Manag. 302, 372–378 (2013).
Pedersen, B. S. The Role of Stress in the Mortality of Midwestern Oaks as Indicated by Growth prior to Death. Ecology 79, 79–93 (1998).
Brum, M. et al. Hydrological niche segregation defines forest structure and drought tolerance strategies in a seasonal Amazon forest. J. Ecol. (2018) doi:10.1111/1365-2745.13022.
Bourdier, T. et al. Tree Size Inequality Reduces Forest Productivity: An Analysis Combining Inventory Data for Ten European Species and a Light Competition Model. PLOS ONE 11, e0151852 (2016).
Niinemets, Ü. Responses of forest trees to single and multiple environmental stresses from seedlings to mature plants: Past stress history, stress interactions, tolerance and acclimation. For. Ecol. Manag. 260, 1623–1639 (2010).
Stephenson, N. L. et al. Rate of tree carbon accumulation increases continuously with tree size. Nature 507, 90–93 (2014).
Ellenberg, H., Weber, H. E., Düll, R., Wirth, V. & Werner, W. Zeigerwerte von Pflanzen in Mitteleuropa. (Verlag Erich Goltze KG, Göttingen, Germany, 2001).
Leuschner, C. & Meier, I. C. The ecology of Central European tree species: Trait spectra, functional trade-offs, and ecological classification of adult trees. Perspect. Plant Ecol. Evol. Syst. 33, 89–103 (2018).
Lübbe, T., Schuldt, B. & Leuschner, C. Acclimation of leaf water status and stem hydraulics to drought and tree neighbourhood: alternative strategies among the saplings of five temperate deciduous tree species. Tree Physiol. 37, 456–468 (2017).
Engelmann, R. A. et al. Der Gehölzbestand des Stieleichen-Ulmen-Hartholzauenwalds (Querco-Ulmetum minoris ISSLER 1942) im Projektgebiet “Lebendige Luppe“ in der nordwestlichen Elster-Luppe-Aue bei Leipzig. UFZ-Ber. 115–132 (2022).
Darnstaedt, F. et al. Gehölzbestand Der Strauchschicht Im Stieleichen-Ulmen-Hartholzauenwald (Querco-Ulmetum Minoris ISSLER 1942) Im Projekt-Gebiet Lebendige Luppe in Der Elster-Luppe-Aue Bei Leipzig. (2024).
Koop, H. Vegetative reproduction of trees in some European natural forests. Vegetatio 72, 103–110 (1987).
Henkel, S. et al. Veränderungen Des Baumbestandes Des Stieleichen-Ulmen- Hartholzauenwaldes (Querco-Ulmetum Minoris) in Der Leipziger Elster-Luppe-Aue. (2024).
Wiegand, J. Development and composition of the regeneration layer of the floodplain forest of Leipzig in dependence on environmental conditions, interspecific competition and the overstorey. (Universität Leipzig, Leipzig, 2021).
Arndt, E., Bernhard, D., Jesche, C., Kupillas, S. & Voigt, W. Species diversity and tree association of Heteroptera (Insecta) in the canopy of a Quercus-Fraxinus-Tilia floodplain forest. in The Canopy of a Temperate Floodplain Forest - First results from 5 years of research at the Leipzig Canopy Crane (eds. Unterseher, M., Morawetz, W., Klotz, S. & Arndt, E.) 81–90 (Universität Leipzig, Merkur, Leipzig, 2007).
Schmidt, C., Bernhard, D. & Arndt, E. Ecological examinations concerning xylobiontic Coleoptera in the canopy of a Quercus-Fraxinus forest. in The Canopy of a Temperate Floodplain Forest - First results from 5 years of research at the Leipzig Canopy Crane (eds. Unterseher, M., Morawetz, W., Klotz, S. & Arndt, E.) 97–105 (Universität Leipzig, Merkur, Leipzig, 2007).
Stenchly, K., Bernhard, D. & Finch, O.-D. Arboricolous spiders (Arachnida, Araneae) of the Leipzig floodplain forest – first results. in The Canopy of a Temperate Floodplain Forest - First results from 5 years of research at the Leipzig Canopy Crane (eds. Unterseher, M., Morawetz, W., Klotz, S. & Arndt, E.) 72–80 (Universität Leipzig, Merkur, Leipzig, 2007).
Unterseher, M., Reiher, A., Otto, P. & Schnittler, M. Pilze im Kronenraum lebender Bäume – 5 Jahre mykologische Biodiversitätsforschung am Leipziger Auwald Kran. Z. Für Mykol. 74, 203–220 (2008).
Haack, N. et al. Patterns of richness across forest beetle communities—A methodological comparison of observed and estimated species numbers. Ecol. Evol. 11, 626–635 (2020).
Hahn, L., Brunk, I., Haack, N. L., Preuss, L. & Bernhard, D. Die Diversität der Coleoptera im Leipziger Auwald – erste Ergebnisse einer mehrjährigen Untersuchung mit Flugfensterfallen im Kronenraum und in der Strauchschicht. Entomol. Nachrichten Berichte 66, 69–89 (2022).
European Environment Agency. European Forest Ecosystems: State and Trends. (Publications Office, LU, 2016).
Petsch, D. K., Cionek, V. de M., Thomaz, S. M. & dos Santos, N. C. L. Ecosystem services provided by river-floodplain ecosystems. Hydrobiologia (2022) doi:10.1007/s10750-022-04916-7.
Scholz, M. et al. Ökosystemfunktionen von Flussauen. BfN Naturschutz Biol. Vielfalt 124, (2012).
Evans, M. R. Will natural resistance result in populations of ash trees remaining in British woodlands after a century of ash dieback disease? R. Soc. Open Sci. 6, 190908 (2019).
McKinney, L. V. et al. The ash dieback crisis: genetic variation in resistance can prove a long-term solution. Plant Pathol. 63, 485–499 (2014).
Pautasso, M., Aas, G., Queloz, V. & Holdenrieder, O. European ash (Fraxinus excelsior) dieback – A conservation biology challenge. Biol. Conserv. 158, 37–49 (2013).
Elles, L. et al. Die Entwicklung Der Baumartenzusammensetzung Im Leipziger Auwald in Abhängigkeit Vom Grundwasserspiegel: Prognosen Eines Demografischen Waldmodells. (2024).

No competing interests reported.

Henkeletal.SupplementaryMaterial.pdf

Download PDF

Editorial decision: Revision requested
03 Oct, 2024
Reviews received at journal
30 Sep, 2024
Reviews received at journal
26 Sep, 2024
Reviews received at journal
24 Sep, 2024
Reviews received at journal
18 Sep, 2024
Reviewers agreed at journal
16 Sep, 2024
Reviewers agreed at journal
16 Sep, 2024
Reviewers agreed at journal
15 Sep, 2024
Reviewers agreed at journal
14 Sep, 2024
Reviewers agreed at journal
13 Sep, 2024
Reviewers agreed at journal
13 Sep, 2024
Reviewers agreed at journal
12 Sep, 2024
Reviewers invited by journal
12 Sep, 2024
Editor assigned by journal
12 Sep, 2024
Editor invited by journal
12 Jul, 2024
Submission checks completed at journal
10 Jul, 2024
First submitted to journal
02 Jul, 2024

You are reading this latest preprint version

Floodplain forests under stress: how ash dieback and hydrology affect tree growth patterns under climate change

Status:

Version 1

Abstract

Figures

1. Introduction

2. Material and Methods

2.1. Study site & design

2.2. Hydrological data

2.3. Tree data

2.4. Data on ash dieback

2.5. Data Analysis

2.5.1. Data preparation

2.5.2. Statistical Analysis

3. Results

3.1. Descriptive statistics

3.2. Upper canopy model

3.3. Lower canopy model

4. Discussion

4.1. General growth responses to ash dieback and effect of hydrological conditions

4.2. Growth responses in the lower and upper canopy

4.3. Species specific growth responses in the light of shade and drought tolerance

4.4. Implications for conservation

5. Conclusion and Outlook

Declarations

Competing interests

Additional information

Author Contribution

Acknowledgement

Data Availability

References

Additional Declarations

Supplementary Files

Status:

Version 1