|
Deep learning for tree line ecotone mapping from remote sensing data
Dvořák, Jakub ; Potůčková, Markéta (advisor) ; Lefèvre, Sébastien (referee)
Deep learning is growing in popularity in the remote sensing community, especially as a classification algorithm. First part of this thesis describes deep neural networks commonly used for remote sensing classification and their various applications. Capabilities of selected geospatial software suites in relation to deep models are also discussed in this part. Theoretical findings from the first part of the thesis are validated using two deep convolutional Encoder-Decoder networks - U-Net and its proposed adaptation called KrakonosNet. They are used to perform a sematic segmentation of spruce trees and dwarf pine shrubs in the tree line ecotone of the Krkonoše Mountains, Czechia. A normalised digital surface model is employed for creation of sufficiently large amount of training data, while the classification itself is performed using only optical imagery with very high spatial resolution. Resulting classification is compared to a set of traditional remote sensing classifiers, namely Maximum Likelihood, Random Forest, and a Support Vector Machine. Both U-Net and KrakonosNet significantly outperform the other classifiers on this dataset and will be consequently used in a related research project. Key words deep learning, U-Net, Krkonoše mountains, classification, vegetation mapping, picea abies,...
|
|
Zhodnocení přirozeného generativního zmlazení borovice kleče v NPR Praděd (CHKO Jeseníky)
Pálka, Ondřej
This diploma thesis deals with the study of the current capability of the generative reproduction of dwarf pine stands (Pinus mugo Turra) in NNR Praděd and the determination of factors influencing the success of generative reproduction. The work deals with the occurrence of the dwarf pine seedlings depending on the stand disturbance of grass-herb vegetation, distance from parental stand and the thickness of the soil horizon A. The work also solves the problem of influence of a particular species of grass-herb vegetation on the occurrence of seedlings. The results show that the higher the rate of disturbance and the smaller the distance from parental stand, the higher is the occurrence of seedlings. It was also found that the most common types of vegetation in which the seedlings were grown were Nardus stricta and Avenella flexuosa.
|
|
Citlivost letokruhových řad borovice kleče a smrku ztepilého ke klimatickým parametrům horní hranice lesa
Mikušková, Jana
Trees growing on the treeline can serve as a sensitive indicator of changes in climatic parameters associated with ongoing climate change. Understanding differences between trees and shrubs growing in this ecotone is a key for understanding the functioning of the treeline. Therefore, the submitted thesis deals with the study climate diameter-growth responseof Norway spruce and Dwarf pine. Samples from Norway spruce and Dwarf pine were collected at the localities at treeline in the National Nature Reserve Praděd in the Jeseníky PLA. Indexed chronologies were developed for Norway spruce and Dwarf pine using various detrending methods of raw tree ring series (Spline, Negative Exponential, Ar, Mean, Friedman). For subsequent dendroclimatological analysis, I used the chronology with their tree ring series were detrended using Spline. Subsequently, correlations of the indexed tree-rings width with monthly climatic variables (average temperature, precipitation total, PDSI – Palmer drough severity index) were calculated. These correlations were investigated both for the entire chronology period and separately for the sliding time sequences within the chronology. It was found that the radial growth of Norway spruce is influenced mainly by temperature in the growing season (May to July) and in November in the year preceding the creation of the tree-ring. In the case of monthly precipitation, their positive correlation with radial growth was recorded only in March and no significant correlations were noted for PDSI. The growth of Dwarf pine is influenced by the temperature at the beginning of the growing season (June) and in October in the year preceding the growth of the tree-ring. Monthly precipitation negatively affects radial growth in October in the year preceding the growth of the tree-ring and no significant correlations were noted for PDSI. The results correspond to the view that the most important factor for the growth of trees at the treeline is temperature, that is not so important for shrubs. The monthly precipitation sums and drought index are not limiting for woody plants of this ecotone.
|
|
Radiální růst borovice kleče na horní hranici lesa v CHKO Jeseníky
Jakubcová, Vanda
This bachelor thesis deals with the study of the reaction of Dwarf pine to the climatic variables at the treeline ecotone. As a representative of the shrubs, the Dwarf pine (Pinus mugo Turra) was selected. At the site of Šerák-Keprník National Nature Reserve in the Protected Landscape Area of Jeseníky, wood discs from 32 individuals of Dwarf pine were taken. The discs surface were grinded and the tree ring widths were measured in the laboratory. The average age of Dwarf pine shrubs in a given location is between 70-80 years. Using 4 detrending methods (negative exponential, mean, friedman and RCS), the age trend was removed from tree ring series, and site chronology was created from the each type of detrending standardization. The resulting chronologies were correlated with climatic data (monthly temperatures, precipitation and Palmer's Drought Index) and the climate signal was analysed.. It was found that the diametr growth of Dwarf pine shrubs in a studied locality is mainly affected by the temperature during the growing season (May, June, July). The precipitation affects the growth of the Dwarf pine in the locality, especially in the autumn months (September, October). PDSI has a rather negative effect on the growth of the dwarf pine shrubs.
|
|
The effect of vegetation pattern on stand expansion on the example of dwarf pine
Růžičková, Veronika ; Treml, Václav (advisor) ; Lipský, Zdeněk (referee)
This bachelor thesis deals with the influence of the structure of the dwarf pine on its expansion in the highest parts of the Giant Mountains. The aim was to find out how this and other factors influence the expansion. For this purpose, in 68 squares, the kneeling vegetation was orchestrated on the basis of orthoctified aerial photographs from 1964 and 2015. Subsequently, a change in the growth of the dwarf pine, the length of the edges of the forest in 1964, the average slope and the altitude in each square were found. Using the correlation analysis, the effect of the variables on the change of the surface was determined. These variables were further used to build a multiple linear regression model. The results showed that the dwarf pine growth increased in the period under review. The greatest influence on expansion is the length of the edges in 1964, ie the breakthrough, the impact of which is positive. A similarly strong influence is the slope of the surface, which affects the expansion negatively. Similar results have been achieved by other authors
|
|
Klimatický signál v letokruhových chronologiích borovice kleče
Samusevich, Alina ; Treml, Václav (advisor) ; Vejpustková, Monika (referee)
Pinus mugo Turra (sensu lato) is a prostrate shrub growing above the alpine timberline in the mountain ranges of Southern, Central and Eastern Europe. It is well adapted to the harsh alpine environment creating different mechanisms for survival. The research was carried out in Krkonoše Mountains on base of Pinus mugo individuals compared with Picea abies trees. Samples from shrubs were gathered using serial sectioning from four sites in different elevations located on Sněžka Mountain and Smogornia Ridge. Further analysis of the sampled material was made with the help of different detrending methods to see which method will return the best growth response to climate parameters. RCS detrending and detrending via simple averaging reflected the best climate signal contained in dwarf pine chronologies. The advantage of these methods is based on their ability to reflect growth conditions of the particular site. Upper sites showed significant correlations with temperatures of the growing season, while on lower sites the signal was quite unclear. The amount of precipitation plays the significant role on shrub growth during the vegetation period (especially July month) and early spring when water is very important for growth initiation. Upper Pinus mugo sites showed high sensitivity to droughts, especially...
|
|
Mapování současného stavu vegetace při horní hranici lesa na území NPR Praděd
Vršanová, Marie
The aim of the thesis was to derive complex information of the current vegetation cover in the alpine treeline ecotone in Praděd Reserve by using remote sensing. Particularly thesis focuses on an area and expansion of dwarf pine (Pinus mugo Turra) at site above the alpine treeline ecotone around the peak Praděd. Two types of aerial photograph from 2000 and 2012 and satellite images from Landsat 7 (yaer 2000) and Landsat 8 (year 2013) were used for mapping vegetation. Supervised and unsupervised classification accesses and also "per-pixel" and objected classification methods were used in image processing. All the classifications confirmed expansion of dwarf pine stands in the alpine treeline vegetation. Most precise method was based on supervised objected classification.
|
| |
| |