12 results on '"Kollert, Andreas"'
Search Results
2. Exploring the potential of land surface phenology and seasonal cloud free composites of one year of Sentinel-2 imagery for tree species mapping in a mountainous region
- Author
-
Kollert, Andreas, Bremer, Magnus, Löw, Markus, and Rutzinger, Martin
- Published
- 2021
- Full Text
- View/download PDF
3. Downscaling MODIS NDSI to Sentinel-2 fractional snow cover by random forest regression
- Author
-
Kollert, Andreas (author), Mayr, Andreas (author), Dullinger, Stefan (author), Hülber, Karl (author), Moser, Dietmar (author), Lhermitte, S.L.M. (author), Gascoin, Simon (author), Rutzinger, Martin (author), Kollert, Andreas (author), Mayr, Andreas (author), Dullinger, Stefan (author), Hülber, Karl (author), Moser, Dietmar (author), Lhermitte, S.L.M. (author), Gascoin, Simon (author), and Rutzinger, Martin (author)
- Abstract
Imagery acquired by the Moderate-resolution Imaging Spectroradiometer (MODIS) provides a global archive of dailyNormalized Difference Snow Index (NDSI) at 500 m nominal resolution since the year 2000. While Sentinel-2 (S2) NDSI provides an increased spatial resolution of 20 m since the year 2015, the temporal resolution amounts to only 5 days and thus lacks the high temporal resolution of MODIS. Efforts to combine NDSI datasets for an increased temporal and spatial resolution have so far focused on the deriving binary snow cover maps or combining data from other sensors. In contrast, we produce fine scale (20 m) fractional snow cover (FSC) by downscaling MODIS NDSI to S2 resolution. Random forest regression predicts S2 NDSI based on dynamic features (MODIS NDSI, day-of-year) and static, topographic features for an alpine study site. Subsequently, FSC is derived from S2 NDSI. Cross-validation results in R2 of 0.795 and RMSE of 0.155 for FSC and outperforms common resampling methods. Multi-annual S2 NDSI metrics are able to slightly improve model accuracy. Our results suggest that combining topographical data and low-resolution NDSI allows to produce daily, high-resolution S2 NDSI and FSC and improve fine scale characterization of snow cover dynamics in mountain landscapes., Mathematical Geodesy and Positioning
- Published
- 2024
- Full Text
- View/download PDF
4. Limited impact of microtopography on alpine plant distribution
- Author
-
Chytrý, Kryštof, primary, Helm, Norbert, additional, Hülber, Karl, additional, Moser, Dietmar, additional, Wessely, Johannes, additional, Hausharter, Johannes, additional, Kollert, Andreas, additional, Mayr, Andreas, additional, Rutzinger, Martin, additional, Winkler, Manuela, additional, Pauli, Harald, additional, Saccone, Patrick, additional, Paetzolt, Mariana, additional, Hietz, Peter, additional, and Dullinger, Stefan, additional
- Published
- 2023
- Full Text
- View/download PDF
5. Limited impact of microtopography on alpine plant distribution.
- Author
-
Chytrý, Kryštof, Helm, Norbert, Hülber, Karl, Moser, Dietmar, Wessely, Johannes, Hausharter, Johannes, Kollert, Andreas, Mayr, Andreas, Rutzinger, Martin, Winkler, Manuela, Pauli, Harald, Saccone, Patrick, Paetzolt, Mariana, Hietz, Peter, and Dullinger, Stefan
- Abstract
Complex topography regulates near‐surface temperature above the treeline. It may thus sustain microrefugia for alpine plants and relax the need of shifting upward when the climate warms. The effectiveness of these microrefugia rests on the premise that plant distributions in alpine landscapes are mainly controlled by fine‐scale topographic variation. We tested this assumption by relating the distribution of 79 plant species and 10 community attributes across 900 1 m² plots in a landscape spanning 1677 m of elevation to 17 topographical descriptors at resolutions between 1 and 301 m. We found that the presence of most species and most community attributes were better explained by topographic variation at coarser scales (> 20 m). Fine‐scale topography is more clearly reflected in moisture than in temperature requirements of species. The elevational gradient rather than topographic variation at any scale, is the single most important driver of both species distributions and the variation in community attributes in the area studied. We hypothesise that our results reveal a hitherto underestimated influence of spatial mass effects on alpine plant distributions. These effects can override environmental filtering at fine scales and will thus impede the survival of cold‐adapted plants in small and fragmented refugia under climate warming. [ABSTRACT FROM AUTHOR]
- Published
- 2024
- Full Text
- View/download PDF
6. Validation of high resolution remotely sensed and modeled snow cover with webcam imagery
- Author
-
Kollert, Andreas, primary, Mayr, Andreas, additional, Rutzinger, Martin, additional, and Dullinger, Stefan, additional
- Published
- 2023
- Full Text
- View/download PDF
7. High-resolution sensing of alpine vegetation location properties by multi-source earth observation techniques
- Author
-
Rutzinger, Martin, primary, Kollert, Andreas, additional, Mayr, Andreas, additional, Müller, Lukas, additional, Hiebl, Benedikt, additional, Bremer, Magnus, additional, and Dullinger, Stefan, additional
- Published
- 2022
- Full Text
- View/download PDF
8. Fine‐scale alpine plant community assembly: Relative roles of environmental sorting, dispersal processes and species interactions.
- Author
-
Helm, Norbert, Chytrý, Kryštof, Hülber, Karl, Moser, Dietmar, Wessely, Johannes, Gattringer, Andreas, Hausharter, Johannes, Pauli, Harald, Winkler, Manuela, Saccone, Patrick, Lamprecht, Andrea, Rutzinger, Martin, Mayr, Andreas, Kollert, Andreas, and Dullinger, Stefan
- Subjects
- *
GLOBAL warming , *PLANT communities , *TIMBERLINE , *PLANT dispersal , *MOUNTAIN plants - Abstract
Besides environmental sorting, other processes like biotic interactions and dispersal limitation are vital for the assembly of plant communities in high mountains and their re‐assembly under changing climatic conditions. Nevertheless, studies that compare the impact of these factors on plant community assembly above the tree line are largely lacking so far. We analysed occurrence changes in vascular plant communities of 492 permanent 1‐m2 plots in the alpine‐nival ecotone of Mt. Schrankogel, Austrian Alps by comparing resurvey data from 2014 with data from the initial survey in 1994. We combined these data with species inventories from 899 additional plots sampled in 2021 and 2022 across a larger landscape above the tree line covering an elevational range of 1700 m, which we used for fine‐scale habitat suitability modelling. We assessed the relative effects of projected habitat suitability, propagule pressure from surrounding populations and biomass density of neighbours on 1532 colonization and 372 extirpation events of 31 species observed on the permanent plots. We found that all three factors are significantly related to both colonisations and extirpations, with habitat suitability having the strongest, propagule pressure a slightly weaker, and vegetation density the weakest effect. Colonisations can be better explained by the three process proxies than extirpations. Our results indicate a crucial role of dispersal limitation besides the predominant effect of environmental filtering on the (re‐)assembly of the alpine‐nival plant community, while competitive/facilitative effects between plants tend to play a minor role. The strong imprint of nearby source plant populations on colonization/extirpation events suggests that recent plant migrations predominantly occur in small steps. This implies that while the topographically complex alpine terrain offers climatic microrefugia for plants, it may also pose potential barriers, hindering species from following their suitable climatic niches upwards. Synthesis: Besides filtering by environmental conditions dispersal limitation had a strong effect on the observed changes in a local alpine plant community over two decades. Limited dispersal capacities of plant species may counteract the ability of isolated cold areas to effectively shelter high alpine plants from the effects of climate warming. [ABSTRACT FROM AUTHOR]
- Published
- 2024
- Full Text
- View/download PDF
9. Stand scale reconstruction of architectural tree models from unmanned aerial vehicle laser scanning (ULS) data.
- Author
-
Bremer, Magnus, Rutzinger, Martin, and Kollert, Andreas
- Published
- 2019
10. Mapping tree species using Sentinel-2 time series : aspects of comprehensive mapping approaches in mountainous environments
- Author
-
Kollert, Andreas Martin and Kollert, Andreas Martin
- Abstract
Satellitenbilder mit einer hohen zeitlichen und räumlichen Auflösung (z.B. Sentinel-2) sind in den letzten Jahren zunehmend frei verfügbar. Trotz des vereinfachten Datenzugangs existieren wenige Ansätze, die eine flächendeckende Kartierung von Baumarten anstreben. Die vorliegende Arbeit beschäftigt sich daher mit der flächendeckenden Kartierung von Baumarten basierend auf Satellitenbildern. Aufbauend auf Sentinel-2-Zeitreihen für das Jahr 2018, die einen Großteil von Tirol (Österreich) abdecken, werden verschiedene methodische Ansätze unter Berücksichtigung topographischer Parameter untersucht.Existierende Ansätze bauen oftmals auf einer multitemporalen Datengrundlage auf. Um die Möglichkeiten zeitlich hochaufgelöster Zeitreihen zu untersuchen, werden in der vorliegenden Arbeit zudem weitere Variablen wie beispielweise zeitlich aggregierte Bilder und phänologische Parameter abgeleitet. Die Güte der Klassifikation wird mittels einer räumlichen Kreuzvalidierung quantifiziert, um den Einfluss der räumlichen Autokorrelation auf die Genauigkeitsmaße gering zu halten. Die Arbeit wird in einem Gebirgsgebiet (Ostalpen, Tirol, Österreich) durchgeführt und untersucht, ob speziell topographische Parameter (Hangneigung, Ausrichtung, Schlagschatten) eine Auswirkung auf die Genauigkeit der abgeleiteten Karten haben. Zudem wird der Einfluss struktureller Waldparameter (Überschirmungsgrad und Baumhöhe) untersucht. Die wichtigsten Aufnahmezeitpunkte und Bänder werden mit Hilfe zusätzlicher Ausgaben des Random Forest Algorithmus eingeschätzt.Die Ergebnisse zeigen, dass fünf verschiedene Baumarten und -typen (Laubbäume, Lärchen, Kiefern, Fichten/Tannen, Bergkiefern) mit einer Genauigkeit von rund 85% kartiert werden können. Phänologische Parameter verbessern die Qualität der Kartierung nicht. Bereiche innerhalb von Schlagschatten und mit nördlicher Ausrichtung zeigen eine höhere Rate von Fehlklassifikationen. Ob Faktoren wie Baumhöhe und Überschirmungsgrad einen Einfluss auf die Klass, This work represents a methodological investigation into the field of accurate and comprehensive mapping of tree species in mountainous environments based on time series of satellite imagery which - despite the nowadays huge amount of freely available imagery - has recently still been regarded as a challenge. The aim is to quantify the value of these both temporally and spatially dense time series, with a special focus on errors that might be related to characteristics of mountainous environments. The study is based on Sentinel-2 time series of the year 2018 that cover a vast part of the Province of Tyrol, Austria.In addition to single Sentinel-2 acquisitions, temporal image composites and phenological metrics are computed and incorporated into the classification workflow in order to evaluate potential benefits. Accuracy assessment is based on spatial cross-validation, a technique that aims at reducing positively biased results due to spatial autocorrelation. Errors are qualitatively analysed with respect to topographic parameters (aspect, slope, shadows cast by mountains) and structural forest parameters (mean tree height, crown coverage). Additional information of the Random Forest algorithm is used to analyse the importance of features with respect to the wavelength and acquisition date.The results show that mapping five different species (broadleaved-, larch-, pine-, spruce/fir- and dwarf pine stands) is possible with an overall accuracy of around 85%. Phenological metrics did not improve mapping accuracies. Areas within cast shadows and northern sectors showed worse results. The influence of stand characteristics (crown coverage, tree height) on mapping accuracy is ambiguous and needs further investigation. It could be shown that many existing approaches cannot be applied in area-wide mapping efforts because they rely on cloud-free imagery which often is not available. Therefore specific preprocessing (gap-filling, temporal aggregation, extracting temporal metr, submitted by Andreas Kollert, Zusammenfassung in deutscher Sprache, Druck-Ausgabe ohne Berücksichtigung der Errata (siehe Digitale Bibliothek), Masterarbeit University of Innsbruck 2019
11. Mapping tree species using Sentinel-2 time series : aspects of comprehensive mapping approaches in mountainous environments
- Author
-
Kollert, Andreas Martin and Kollert, Andreas Martin
- Abstract
Satellitenbilder mit einer hohen zeitlichen und räumlichen Auflösung (z.B. Sentinel-2) sind in den letzten Jahren zunehmend frei verfügbar. Trotz des vereinfachten Datenzugangs existieren wenige Ansätze, die eine flächendeckende Kartierung von Baumarten anstreben. Die vorliegende Arbeit beschäftigt sich daher mit der flächendeckenden Kartierung von Baumarten basierend auf Satellitenbildern. Aufbauend auf Sentinel-2-Zeitreihen für das Jahr 2018, die einen Großteil von Tirol (Österreich) abdecken, werden verschiedene methodische Ansätze unter Berücksichtigung topographischer Parameter untersucht.Existierende Ansätze bauen oftmals auf einer multitemporalen Datengrundlage auf. Um die Möglichkeiten zeitlich hochaufgelöster Zeitreihen zu untersuchen, werden in der vorliegenden Arbeit zudem weitere Variablen wie beispielweise zeitlich aggregierte Bilder und phänologische Parameter abgeleitet. Die Güte der Klassifikation wird mittels einer räumlichen Kreuzvalidierung quantifiziert, um den Einfluss der räumlichen Autokorrelation auf die Genauigkeitsmaße gering zu halten. Die Arbeit wird in einem Gebirgsgebiet (Ostalpen, Tirol, Österreich) durchgeführt und untersucht, ob speziell topographische Parameter (Hangneigung, Ausrichtung, Schlagschatten) eine Auswirkung auf die Genauigkeit der abgeleiteten Karten haben. Zudem wird der Einfluss struktureller Waldparameter (Überschirmungsgrad und Baumhöhe) untersucht. Die wichtigsten Aufnahmezeitpunkte und Bänder werden mit Hilfe zusätzlicher Ausgaben des Random Forest Algorithmus eingeschätzt.Die Ergebnisse zeigen, dass fünf verschiedene Baumarten und -typen (Laubbäume, Lärchen, Kiefern, Fichten/Tannen, Bergkiefern) mit einer Genauigkeit von rund 85% kartiert werden können. Phänologische Parameter verbessern die Qualität der Kartierung nicht. Bereiche innerhalb von Schlagschatten und mit nördlicher Ausrichtung zeigen eine höhere Rate von Fehlklassifikationen. Ob Faktoren wie Baumhöhe und Überschirmungsgrad einen Einfluss auf die Klass, This work represents a methodological investigation into the field of accurate and comprehensive mapping of tree species in mountainous environments based on time series of satellite imagery which - despite the nowadays huge amount of freely available imagery - has recently still been regarded as a challenge. The aim is to quantify the value of these both temporally and spatially dense time series, with a special focus on errors that might be related to characteristics of mountainous environments. The study is based on Sentinel-2 time series of the year 2018 that cover a vast part of the Province of Tyrol, Austria.In addition to single Sentinel-2 acquisitions, temporal image composites and phenological metrics are computed and incorporated into the classification workflow in order to evaluate potential benefits. Accuracy assessment is based on spatial cross-validation, a technique that aims at reducing positively biased results due to spatial autocorrelation. Errors are qualitatively analysed with respect to topographic parameters (aspect, slope, shadows cast by mountains) and structural forest parameters (mean tree height, crown coverage). Additional information of the Random Forest algorithm is used to analyse the importance of features with respect to the wavelength and acquisition date.The results show that mapping five different species (broadleaved-, larch-, pine-, spruce/fir- and dwarf pine stands) is possible with an overall accuracy of around 85%. Phenological metrics did not improve mapping accuracies. Areas within cast shadows and northern sectors showed worse results. The influence of stand characteristics (crown coverage, tree height) on mapping accuracy is ambiguous and needs further investigation. It could be shown that many existing approaches cannot be applied in area-wide mapping efforts because they rely on cloud-free imagery which often is not available. Therefore specific preprocessing (gap-filling, temporal aggregation, extracting temporal metr, submitted by Andreas Kollert, Zusammenfassung in deutscher Sprache, Druck-Ausgabe ohne Berücksichtigung der Errata (siehe Digitale Bibliothek), Masterarbeit University of Innsbruck 2019
12. 4D Forest moisture mapping based on multi-temporal earth observation signatures (4D-FORMAT).
- Author
-
Rutzinger, Martin, Bremer, Magnus, Kollert, Andreas, Hollaus, Markus, Pfeifer, Norbert, Bruggisser, Moritz, Bauer-Marschallinger, Bernhard, Hagen, Karl, Schadauer, Klemens, Gartner, Karl, and Bauerhansl, Christoph
- Published
- 2018
Catalog
Discovery Service for Jio Institute Digital Library
For full access to our library's resources, please sign in.