Refine
Document Type
- Doctoral Thesis (3)
- Article (2)
Language
- English (5)
Has Fulltext
- yes (5)
Is part of the Bibliography
- no (5)
Keywords
- wood anatomy (5) (remove)
Publisher
Dendrochronology, the science of tree-rings is a tool which has been widely used for many years for understanding changes in the environment, as trees react to environmental changes over time. In the contemporary situation, where climate warming in the Arctic is unequivocal and its effects on the Alpine and tundra ecosystems are seen pronouncedly in the past decade, the role of dendro-studies and the use of trees and shrubs alike as proxies of change has become critical. Studies clearly indicate that warming in the Arctic and Alpine tundra has resulted in increased vegetation in recent years. Shrubs, in these sensitive ecosystems, have proven to be highly instrumental as they likely benefit from this warming and hence are good indicators and auditees of this change. Therefore, in this study, we investigate the potential of shrubs in the evolving field of dendro-ecology/climatology.
Studies from classical dendrochronology used annual rings from trees. Further, because of shrub sensitivity to contemporary change, shrub-based dendrochronological research has increased at a notable scale in the last decade and will likely continue. This is because shrubs grow even beyond the tree line and promise environmental records from areas where tree growth is very limited or absent. However, a common limitation noted by most shrub studies is the very hard cross-dating due to asynchronous growth patterns. This limitation poses a major hurdle in shrub-based dendrochronological studies, as it renders weak detection of common signals in growth patterns in population stands. This common signal is traced by using a ‘site-chronology’.
In this dissertation, I studied shrub growth through various resolutions, starting from understanding radial growth within individuals along the length of the stem, to comparison of radial growth responses among male and female shrubs, to comparing growth responses among trees and shrubs to investigation of biome-wide functional trait responses to current warming. Apart from Chapter 4 and Chapter 6, I largely used Juniperus communis sp. for investigations as it is the most widely distributed woody dioecious species often used in dendro-ecological investigations in the Northern Hemisphere.
Primarily, we investigated radial growth patterns within shrubs to better understand growth within individuals by comparing different stem-disks from different stem heights within individuals. We found significant differences in radial growth from different stem-disks with respect to stem heights from same individuals. Furthermore, we found that these differences depending on the choice of the stem-disk affect the resulting site-chronology and hence climate-sensitivity to a substantial extent and that the choice of a stem-disk is a crucial precursor which affects climate-growth relationships.
Secondly, we investigated if gender difference – often reported causing differential radial growth in dioecious trees – is an influential factor for heterogeneous growth. We found that at least in case of Juniperus communis. L and Juniperus communis ssp nana. WILLD there is no substantial gender biased difference in radial growth which might affect the site-chronology. We did find moderate differences between sexes in an overall analysis and attribute this to reproductive effort in females.
In our study to test the potential of shrubs for reconstruction, we used a test case of Alnus viridis ssp crispa. We found a strong correlation between ring-width indices and summer temperature. Initially, the model failed the stability tests when we tested the stability of this relation using a response function model. However, using wood-anatomical analysis we discovered that this was because of abnormal cell-wall formation resulting in very thin rings in the year 2004. Pointer year analysis revealed that the thin rings were caused because of a moth larval outbreak and when corrected for these rings the model passed all stability tests.
Furthermore, to see if trees and shrubs growing in same biomes react to environmental changes similarly, a network analysis with sites ranging from the Mediterranean biome to the Ural Mountains in Russia was carried out. We found that shrubs react better to the current climate warming and have a decoupled divergent temperature response as compared to coexisting trees. This outcome reiterated the importance of shrub studies in relation to contemporary climate change. Even though trees and shrubs are woody forms producing annual rings, they have very different growth patterns and need different methods for analysis and data treatment.
Finally, in a domain-wide network analysis from plant-community vegetation survey, we investigated functional relationships between plant traits (leaf area, plant height, leaf nitrogen content, specific leaf area (SLA), and leaf dry matter content (LDMC)) and abiotic factors viz. temperature and soil moisture. We found a strong relation between summer temperature and community height, SLA and LDMC on a spatial scale. Contrarily, the temporal-analysis revealed SLA and LDMC lagged and did not respond to temperature over the last decade. We realized that there are complex interactions between intra-specific and inter-specific plant traits which differ spatially and temporally impacting Arctic ecosystems in terms of carbon turn over, surface albedo, water balance and heat-energy fluxes. We found that ecosystem functions in the Arctic are closely linked with plant height and will be indicative of warming in the short term future becoming key factors in modelling ecosystem projections.
The recent developments in artificial intelligence have the potential to facilitate new research methods in ecology. Especially Deep Convolutional Neural Networks (DCNNs) have been shown to outperform other approaches in automatic image analyses. Here we apply a DCNN to facilitate quantitative wood anatomical (QWA) analyses, where the main challenges reside in the detection of a high number of cells, in the intrinsic variability of wood anatomical features, and in the sample quality. To properly classify and interpret features within the images, DCNNs need to undergo a training stage. We performed the training with images from transversal wood anatomical sections, together with manually created optimal outputs of the target cell areas. The target species included an example for the most common wood anatomical structures: four conifer species; a diffuse-porous species, black alder (Alnus glutinosa L.); a diffuse to semi-diffuse-porous species, European beech (Fagus sylvatica L.); and a ring-porous species, sessile oak (Quercus petraea Liebl.). The DCNN was created in Python with Pytorch, and relies on a Mask-RCNN architecture. The developed algorithm detects and segments cells, and provides information on the measurement accuracy. To evaluate the performance of this tool we compared our Mask-RCNN outputs with U-Net, a model architecture employed in a similar study, and with ROXAS, a program based on traditional image analysis techniques. First, we evaluated how many target cells were correctly recognized. Next, we assessed the cell measurement accuracy by evaluating the number of pixels that were correctly assigned to each target cell. Overall, the “learning process” defining artificial intelligence plays a key role in overcoming the issues that are usually manually solved in QWA analyses. Mask-RCNN is the model that better detects which are the features characterizing a target cell when these issues occur. In general, U-Net did not attain the other algorithms’ performance, while ROXAS performed best for conifers, and Mask-RCNN showed the highest accuracy in detecting target cells and segmenting lumen areas of angiosperms. Our research demonstrates that future software tools for QWA analyses would greatly benefit from using DCNNs, saving time during the analysis phase, and providing a flexible approach that allows model retraining.
Tree growth in northern and upper treeline ecotones of the circumpolar boreal forest is
generally limited by temperature, i.e., trees grow generally more under warm, and less under
cold climatic conditions. Based on the assumption that this relationship between tree growth
and climate is linear and stable through time, dendroclimatologists use tree rings as natural
archives to reconstruct past temperature conditions. Such tree-ring based reconstructions,
together with other natural archives (e.g., ice cores and pollen), constitute our understanding of
past climatic conditions that reach beyond modern instrumental records.
However, a steadily increasing amount of studies reports a recent reduction or loss of the
summer temperature signal for several species and sites of the boreal forest. Such a reduction
of temperature sensitivity results in temporally unstable climate-tree growth relationships,
which challenges the work of dendroclimatologists by potentially leading to miscalibrations of
past climatic conditions. On the upside, this shift in the trees’ climate sensitivity might point to
a shift in tree growth-limiting factors and thus serve as an early indicator of climate change
impacts. There is evidence that this recent reduction in temperature sensitivity might be caused
by the observed strong temperature increase at high latitudes, and thus temperature-induced
drought stress. Other potential drivers and amplifiers of this phenomenon are differing microsite
conditions (dry vs. wet soils) and factors inherent to trees, like genetic properties or age
effects.
In this PhD thesis, I systematically assessed the effects of frequently discussed drivers of
unstable climate-tree growth relationships (climate change, micro-site effects, genetical
predisposition) on two representative species of the boreal forest, white spruce in North
America and Scots pine in Eurasia, across various temporal and spatial scales. I used classical
(tree-ring width) and more novel (wood density, quantitative wood anatomy)
dendrochronological proxies to unravel the effects from annual to sub-monthly resolution.
More precisely, in chapter I, white spruce clones were compared to non-clones at two treeline
sites in Alaska to test whether their growth patterns differ, and whether white spruce clones are
generally suitable for dendroclimatic assessments. Clonal reproduction is frequent at treeline
due to harsh conditions, but might lead to competition among individuals due to the close
proximity among each other, which in turn might obscure their climatic signal. Second, I tested
the effect of warmer and drier climatic conditions on the summer temperature signal of Scots
pine in Eurasia (chapter II) and on the growing season moisture signal of white spruce in North
America (chapter III), respectively. Temperature-induced drought stress is expected to be the
most important driver of unstable climate-growth relationships in the boreal forest. I included
several sites across latitudinal (50-150 km) and longitudinal (1,000-2,200 km) gradients to
cover large parts of the species’ distribution ranges. Since Scots pine covers a wide range of
ecological habitats, I additionally tested the effect of dry and wet micro-site conditions on the
summer temperature signal of Scots pine in chapter II. Finally, in chapter IV, a systematic
literature review was carried out in order to investigate the distribution of unstable climategrowth
relationships in global tree-ring studies, and the usage of such series in climate
reconstructions. Furthermore, the scientific impact of these potentially inaccurate climate
reconstructions was assessed.
In this PhD project, warmer and drier climatic conditions led to temporally unstable climate
signals in both Scots pine (chapter II) and white spruce (chapter III), as expected. Unstable
climate-growth relationships were found for all tested tree-ring proxies and at all sites in North
America, and at most sites in Eurasia. Micro-site effects (chapter II) and clonal growth
(chapter I) had no significant effect on the climate sensitivity and high-frequency variability
of the tested species, but affected absolute growth. The review (chapter IV) revealed that the
phenomenon of unstable climate-growth relationships is globally widespread, and occurs
independent of tree species, geographic location, and tree-ring and climate proxies. While
reconstructions inferred from these unstable relationships are frequent and respective papers
have a high impact, the tree-ring community seems to increasingly recognize the challenge of
unstable climate-growth relationships.
With these findings, this PhD project helped to shed more light on the frequency, underlying
drivers, and the impact of unstable climate-growth relationships in boreal forest trees, as well
as underlying reaction processes in trees. Above all, this PhD project suggests that the loss of
climate sensitivity is caused by a change of growth limiting factors: temperature limitation
seems to be suspended in warmer and drier years for Scots pine in Eurasia, and moisture
limitation first arises under warm/dry conditions for white spruce in North America. Due to
plastic growth responses in trees, the general assumption in dendroclimatology – that climategrowth
relationships are stable through time – seems to be incompatible with the principle of
limiting factors (one factors is always most growth limiting).
To improve the validity of future climate reconstructions, statistical approaches considering
synchronously or changing climatic limiting factors need to be promoted, along with attempts
to select the best responding trees from a dataset. Furthermore, a better understanding of nonclimatic
factors potentially affecting tree growth (e.g., age, disturbance, soil parameters) is
needed. A growth reduction of mature and dominant white spruce trees sampled in this PhD
project seems likely under future warming conditions, with series of wood cells being valuable
early indicators of climate change effects in white spruce. However, inferences cannot be
extended to the entire stand due to the applied sample design. Projected climate warming will
probably lead to a further reduction of the summer temperature signal in trees of the northern
boreal forest, while wider consequences for forest growth and productivity are unclear.
Determining the effect of a changing climate on tree growth will ultimately depend on our understanding of wood formation processes and how they can be affected by environmental conditions. In this context, monitoring intra-annual radial growth with high temporal resolution through point dendrometers has often been used. Another widespread approach is the microcoring method to follow xylem and phloem formation at the cellular level. Although both register the same biological process (secondary growth), given the limitations of each method, each delivers specific insights that can be combined to obtain a better picture of the process as a whole. To explore the potential of visualizing combined dendrometer and histological monitoring data and scrutinize intra-annual growth data on both dimensions (dendrometer → continuous; microcoring → discrete), we developed DevX (Dendrometer vs. Xylogenesis), a visualization application using the “Shiny” package in the R programming language. The interactive visualization allows the display of dendrometer curves and the overlay of commonly used growth model fits (Gompertz and Weibull) as well as the calculation of wood phenology estimates based on these fits (growth onset, growth cessation, and duration). Furthermore, the growth curves have interactive points to show the corresponding histological section, where the amount and development stage of the tissues at that particular time point can be observed. This allows to see the agreement of dendrometer derived phenology and the development status at the cellular level, and by this help disentangle shrinkage and swelling due to water uptake from actual radial growth. We present a case study with monitoring data for Acer pseudoplatanus L., Fagus sylvatica L., and Quercus robur L. trees growing in a mixed stand in northeastern Germany. The presented application is an example of the innovative and easy to access use of programming languages as basis for data visualization, and can be further used as a learning tool in the topic of wood formation and its ecology. Combining continuous dendrometer data with the discrete information from histological-sections provides a tool to identify active periods of wood formation from dendrometer series (calibrate) and explore monitoring datasets.
Forest ecosystems around the world and especially boreal forests, are facing
drastically changing climatic conditions. It is known that these changes could
challenge their functionality and vitality. Still, the exact impact is not fully
understood, as tree growth is a complex process and depends on countless
environmental and genetic factors. To estimate the effects of climate change
on tree growth and forest development precisely, we must learn more about
tree growth itself. A comprehensive approach is needed where trees and
forests are investigated on different scales and levels of detail, ranging from
global studies to studies on single individuals.
In this dissertation, I follow such a comprehensive approach, using the
North American conifer white spruce as an example. I present three papers
in the form of three chapters in which my co-authors and I studied the
growth and anatomy of white spruce (Picea glauca [Moench] Voss) and how
it is influenced by environmental, climatic, and genetic factors.
We used diverse approaches and methods on different spatial scales, ranging from
investigations on the landscape to the local scale. We established three paired
plots with forest and treeline sites (two cold-limited and one drought-limited).
as well as one additional forest site. In the first chapter, we concentrated
on the genetic diversity of white spruce within and between populations at
all study sites throughout Alaska. The genetic investigations were combined
with analyses on the individual growth response of trees to climatic conditions
to find whether genetic similarities or spatial proximity caused similarities
in growth and climatic sensitivity. In the second chapter, we studied the
direct and indirect effects of environmental conditions on the xylem tissue
of white spruce. We analyzed the impact of precipitation, temperature, and
tree height on four xylem anatomical traits in trees growing at the three
treelines. The investigated traits represented the main functions of xylem
tissue (i.e., water transport and structural support). In the third chapter,
we investigated similar xylem anatomical traits at one cold-limited treeline.
We compared xylem anatomy and annual increment between genetic groups
and individuals and between spatial groups to investigate whether spatial or
genetic grouping influenced the anatomy and growth of white spruce.
We found an overall high gene flow and high genetic diversity in white
spruce. However, the sensitivity of the growth and anatomical traits of white
spruce was driven mainly by spatial rather than genetic effects and differed
between study sites. Trees from the drought-limited site were more sensitive
towards precipitation and a moisture index, while trees from the cold-limited
sites were more sensitive towards temperature. A strong direct effect of tem-
perature was primarily found in latewood traits related to the structural sup-
port of the tree. Earlywood traits related to water transport, however, were
influenced mainly by tree height. Tree height itself was potentially affected
by diverse abiotic and biotic factors (e.g., (micro)climate, soil conditions,
and competition). Thus, traits related to water transport were indirectly
influenced by environmental conditions. Genetic effects in xylem anatomical
traits were found in the earlywood hydraulic diameter and latewood den-
sity, whereas in general, primarily spatial rather than genetic grouping was
influencing the anatomy of white spruce.
Overall, white spruce showed to be a genetically diverse species with a
high gene flow. The effects of spatial proximity and spatial grouping on the
sensitivity and anatomy of white spruce indicate high phenotypic plastic-
ity. This high phenotypic plasticity combined with the vast genetic diversity
translates into an immense potential for the species to adjust (phenotypically)
and possibly adapt (genetically) to changing conditions. Thus, in terms of
climate change, white spruce may be a rather persistent species that manages
to cope with the drastic changes. Though additional work might be needed to
draw a more solid conclusion, the presented work shows how a comprehensive
study approach can help to interpret and understand the growth and ecology
of a tree species. It may be an inspiration for future studies to broaden their
approaches and to use comprehensive methods on different levels of detail to
not only observe trees but to explore and understand them.