Description / Table of Contents:
Vegetation change at high latitudes is one of the central issues nowadays with respect to ongoing climate changes and triggered potential feedback. At high latitude ecosystems, the expected changes include boreal treeline advance, compositional, phenological, physiological (plants), biomass (phytomass) and productivity changes. However, the rate and the extent of the changes under climate change are yet poorly understood and projections are necessary for effective adaptive strategies and forehanded minimisation of the possible negative feedbacks.
The vegetation itself and environmental conditions, which are playing a great role in its development and distribution are diverse throughout the Subarctic to the Arctic. Among the least investigated areas is central Chukotka in North-Eastern Siberia, Russia. Chukotka has mountainous terrain and a wide variety of vegetation types on the gradient from treeless tundra to northern taiga forests. The treeline there in contrast to subarctic North America and north-western and central Siberia is represented by a deciduous conifer, Larix cajanderi Mayr. The vegetation varies from prostrate lichen Dryas octopetala L. tundra to open graminoid (hummock and non-hummock) tundra to tall Pinus pumila (Pall.) Regel shrublands to sparse and dense larch forests.
Hence, this thesis presents investigations on recent compositional and above-ground biomass (AGB) changes, as well as potential future changes in AGB in central Chukotka. The aim is to assess how tundra-taiga vegetation develops under changing climate conditions particularly in Fareast Russia, central Chukotka. Therefore, three main research questions were considered:
1) What changes in vegetation composition have recently occurred in central Chukotka?
2) How have the above-ground biomass AGB rates and distribution changed in central Chukotka?
3) What are the spatial dynamics and rates of tree AGB change in the upcoming millennia in the northern tundra-taiga of central Chukotka?
Remote sensing provides information on the spatial and temporal variability of vegetation. I used Landsat satellite data together with field data (foliage projective cover and AGB) from two expeditions in 2016 and 2018 to Chukotka to upscale vegetation types and AGB for the study area. More specifically, I used Landsat spectral indices (Normalised Difference Vegetation Index (NDVI), Normalised Difference Water Index (NDWI) and Normalised Difference Snow Index (NDSI)) and constrained ordination (Redundancy analysis, RDA) for further k-means-based land-cover classification and general additive model (GAM)-based AGB maps for 2000/2001/2002 and 2016/2017. I also used Tandem-X DEM data for a topographical correction of the Landsat satellite data and to derive slope, aspect, and Topographical Wetness Index (TWI) data for forecasting AGB.
Firstly, in 2016, taxa-specific projective cover data were collected during a Russian-German expedition. I processed the field data and coupled them with Landsat spectral Indices in the RDA model that was used for k-means classification. I could establish four meaningful land-cover classes: (1) larch closed-canopy forest, (2) forest tundra and shrub tundra, (3) graminoid tundra and (4) prostrate herb tundra and barren areas, and accordingly, I produced the land cover maps for 2000/2001/2002 and 2016/20017. Changes in land-cover classes between the beginning of the century (2000/2001/2002) and the present time (2016/2017) were estimated and interpreted as recent compositional changes in central Chukotka. The transition from graminoid tundra to forest tundra and shrub tundra was interpreted as shrubification and amounts to a 20% area increase in the tundra-taiga zone and 40% area increase in the northern taiga. Major contributors of shrubification are alder, dwarf birch and some species of the heather family. Land-cover change from the forest tundra and shrub tundra class to the larch closed-canopy forest class is interpreted as tree infilling and is notable in the northern taiga. We find almost no land-cover changes in the present treeless tundra.
Secondly, total AGB state and change were investigated for the same areas. In addition to the total vegetation AGB, I provided estimations for the different taxa present at the field sites. As an outcome, AGB in the study region of central Chukotka ranged from 0 kg m-2 at barren areas to 16 kg m-2 in closed-canopy forests with the larch trees contributing the highest. A comparison of changes in AGB within the investigated period from 2000 to 2016 shows that the greatest changes (up to 1.25 kg m 2 yr 1) occurred in the northern taiga and in areas where land cover changed to larch closed-canopy forest. Our estimations indicate a general increase in total AGB throughout the investigated tundra-taiga and northern taiga, whereas the tundra showed no evidence of change in AGB within the 15 years from 2002 to 2017.
In the third manuscript, potential future AGB changes were estimated based on the results of simulations of the individual-based spatially explicit vegetation model LAVESI using different climate scenarios, depending on Representative Concentration Pathways (RCPs) RCP 2.6, RCP 4.5 and RCP 8.5 with or without cooling after 2300 CE. LAVESI-based AGB was simulated for the current state until 3000 CE for the northern tundra-taiga study area for larch species because we expect the most notable changes to occur will be associated with forest expansion in the treeline ecotone. The spatial distribution and current state of tree AGB was validated against AGB field data, AGB extracted from Landsat satellite data and a high spatial resolution image with distinctive trees visible. The simulation results are indicating differences in tree AGB dynamics plot wise, depending on the distance to the current treeline. The simulated tree AGB dynamics are in concordance with fundamental ecological (emigrational and successional) processes: tree stand formation in simulated results starts with seed dispersion, tree stand establishment, tree stand densification and episodic thinning. Our results suggest mostly densification of existing tree stands in the study region within the current century in the study region and a lagged forest expansion (up to 39% of total area in the RCP 8.5) under all considered climate scenarios without cooling in different local areas depending on the closeness to the current treeline. In scenarios with cooling air temperature after 2300 CE, forests stopped expanding at 2300 CE (up to 10%, RCP 8.5) and then gradually retreated to their pre-21st century position. The average tree AGB rates of increase are the strongest in the first 300 years of the 21st century. The rates depend on the RCP scenario, where the highest are as expected under RCP 8.5.
Overall, this interdisciplinary thesis shows a successful integration of field data, satellite data and modelling for tracking recent and predicting future vegetation changes in mountainous subarctic regions. The obtained results are unique for the focus area in central Chukotka and overall, for mountainous high latitude ecosystems.
Type of Medium:
Dissertation, Potsdam, Universität Potsdam, 2022
1.1 Scientific background
1.2 Study region
1.3 Aims and objectives
2 Materials and methods
3.1 Section 4 - Strong shrub expansion in tundra-taiga, tree infilling in taiga and stable tundra in central Chukotka (north-eastern Siberia) between 2000 and 2017
3.2 Section 5 - Recent above-ground biomass changes in central Chukotka (NE Siberia) combining field-sampling and remote sensing
3.3 Section 6 - Future spatially explicit tree above-ground biomass trajectories revealed for a mountainous treeline ecotone using the individual-based model LAVESI
4 Strong shrub expansion in tundra-taiga, tree infilling in taiga and stable tundra in central Chukotka (north-eastern Siberia) between 2000 and 2017
2 Materials and methods
2.1 Field data collection and processing
2.2 Landsat data, pre-processing and spectral indices processing
2.3 Redundancy analysis (RDA) and classification approaches
3.1 General characteristics of the vegetation field data
3.2 Relating field data to Landsat spectral indices in the RDA model
3.3 Land-cover classification
3.4 Land-cover change between 2000 and 2017
4.1 Dataset limitations and optimisation
4.2 Vegetation changes from 2000/2001/2002 to 2016/2017
Data availability statement
Appendix A. Detailed description of Landsat acquisitions
Appendix B. MODIS NDVI time series from 2000 to 2018
Appendix C. Landsat Indices values for each analysed vegetation site
Appendix D. Fuzzy c-means classification for interpretation of uncertainties for land-cover mapping
Appendix E. Validation of land-cover maps
Appendix F. K-means classification results
Appendix G. Heterogeneity of natural landscapes and mixed pixels of satellite data
Appendix H. Distribution of land-cover classes and their changes by study area
5 Recent above-ground biomass changes in central Chukotka (NE Siberia) combining field-sampling and remote sensing
2 Materials and methods
2.1 Study region and field surveys
2.2 Above-ground biomass upscaling and change derivation
3.1 Vegetation composition and above-ground biomass
3.2 Upscaling above-ground biomass using GAM
3.3 Change of above-ground biomass between 2000 and 2017 in the four focus areas
4.1 Recent state of above-ground biomass at the field sites
4.2 Recent state of above-ground biomass upscaled for central Chukotka
4.3 Change in above-ground biomass within the investigated 15–16 years in central Chukotka
Data availability statement
Appendix A. Sampling and above-ground biomass (AGB) calculation protocol for field data
6 Future spatially explicit tree above-ground biomass trajectories revealed for a mountainous treeline ecotone using the individual-based model LAVESI
2 Materials and methods
2.1 Study region
2.2 LAVESI model setup, parameterisation, and validation
2.2.4 LAVESI simulation setup for this study
2.2.5 Validation of the model’s performance
3.1 Dynamics and spatial distribution changes of tree above-ground-biomass
3.2 Spatial and temporal validation of the contemporary larch AGB
4.1 Future dynamics of tree AGB at a plot level
4.2 What are the future dynamics of tree AGB at the landscape level?
Appendix B. Permutation tests for tree presence versus topographical parameters
Appendix C. Landsat-based, field, and simulated estimations of larch above-ground biomass (AGB).
7.1 What changes in vegetation composition have happened from 2000 to 2017 in central Chukotka?
7.2 How have the above-ground biomass (AGB) distribution and rates changed from 2000 to 2017 in central Chukotka?
7.3 What are the spatial dynamics and rates of tree AGB change in the upcoming centuries in the northern tundra-taiga from 2020 to 3000 CE on the plot level and landscape level?
AWI Reading room