Bachelor

Background: The leaf area index (LAI) is a key parameter for describing vegetation structures and is widely used in models for biophysical processes in ecosystems and vegetation productivity. Many algorithms have been developed for the estimation of LAI based on remote sensing. Here hyperspectral data and LAI measurements will be recorded in the three German Biodiversity Exploratories. The project aims to effectively link hyperspectral data and LAI measurements and identify suitable hyperspectral vegetation indices for LAI estimations in different grassland plots. The fieldwork will take place in spring 2024 and it would be beneficial if the student could (partly) participate in the field survey.

Objectives of this thesis are:

• Take field spectroscopy and LAI measurements in the field in April/Mai 2024
• Apply existing pre-processing functions for hyperspectral data in R
• Calculate a range of vegetation metrics from the hyperspectral data
• Apply regression models to estimate LAI from hyperspectral reflectance data
• Compare the performance of vegetation indices

Requirements:

• Basics in spectral remote sensing
• Interest in hyperspectral data
• Experience in scripting using R and basic regression statistics

Project:

• SensGrün

Contact:

• Christine Wallis
• Birgit Kleinschmit
• Michael Förster

Background: Grassland biomass plays a crucial role in maintaining ecosystem health and supporting various ecological functions. Monitoring grassland biomass is vital for assessing its overall health and productivity. Hyperspectral remote sensing, with its ability to capture detailed spectral information, offers a powerful tool for this task. By analyzing the reflected light from grasslands, hyperspectral sensors can provide precise data on vegetation composition, health, and biomass. This information is promising for land management, conservation efforts, and addressing challenges like climate change and sustainable agriculture. This study aims to model biomass from hyperspectral satellite data in multiple plots of the German biodiversity exploratories.

Objectives of this thesis are:

• Extracting the spectral signal from pre-processed hyperspectral satellite datasets
• Apply regression models to estimate biomass from multiple sites
• Exploring the importance of spectral regions

Requirements:

• Basics in spectral remote sensing
• Interest in hyperspectral data
• Experience in scripting using R and regression statistics

Project:

• SensGrün

Contact:

• Christine Wallis
• Birgit Kleinschmit
• Michael Förster

(Please feel free to contact the contact persons also if you would like to write a Master thesis on this topic)

Background:

One supervisory task of the German federal railway service (Eisenbahnbundesamt) is the inspection of compensation measures (Ausgleichs- und Ersatzmaßnahmen) according to the German federal mitigation regulation (Eingriffsregelung). Instead of expensive and time consuming on-site controls, those monitoring tasks could be potentially covered by time series from the freely available satellite data of the European earth observation program Copernicus. For instance, the Multi-spectral sensors Sentinel-2A and Sentinel-2B are highly sensitive to chlorophyll content and color of vegetation surfaces. Because the Sentinel-2 mission provides images since 2014 with an approximate 5-day temporal resolution, we assume that the 10(-60)m pixel resolved data gives a good baseline for many automated monitoring tasks in the control of nature conservation measures. Potential research questions of the thesis could be: Are long-term trends from afforestation sites measurable? Are breaks from single maintenance measures observable and measurable?

Objectives of this thesis are:

• acquisition of pre-processed Sentinel-2 L2A time series data for multiple test sites
• developing algorithms for the automated detection of trends and breakpoints in S2 backscatter time series (e.g. bfast, trendr, green-brown)
• apply the algorithms on the object-level based on multiple compensation sites

Requirements:

Experience in scripting using R (preferred) or Python

Interest in optical remote sensing and Sentinel-2 MSI data

Expertise in landscape planning, impact assessment and mitigation planning is beneficial

Partners: German Centre for Rail Traffic Research (DZSF) from the Federal Railway Authority (EBA)

Contact:

Christian Schulz (christian.schulz.1(at)tu-berlin.de)

Birgit Kleinschmit (birgit.kleinschmit(at)tu-berlin.de)

Background: Along to the professional species field observations provided by German federal authorities (e.g., Bundeswaldinventur) and research projects (e.g., TERENO, FloraWeb), many alternative sources from Crowd Sourcing platforms (e.g. Pl@ntNet and FloraIncognita) pop up. Those free and large datasets could be very interesting for species distribution mapping, phenological research, and vegetation classification with satellite data. But how is the quality of the data? We search for a self-initiated student who is interested in analytical comparisons of different data sources and different types of geodata.

Objectives: Literature studies on the topic of voluntary geographical information (VGI). Acquisition and filtering the geodata. NDVI pheno profiles comparison of different vegetation types.

Research area: preferably on the national level (Germany) or regional level (Bundesland)
Related to: TreeSatAI
Contact: Christian Schulz

Background

The Harz National Park is the first German National Park covering two federal states and includes complex biotopes over various vegetation zones. It is a significant recreational area but undergoes massive ecological changes that affect the local spruce, beach and mixed forests. Calamities such as bark beetle attack, drought, storm and fire shape the ecosystem transition. Monitoring and understanding such shifts using remote sensing has great potential to rapidly gather supporting information.

In August 2022, a wildfire at the Quesenbank near the village Schierke and the Harzbahn train tracks affected an area of ca. 13 ha. The fire was extinguished successfully, yet its impact on the damaged spruce forest remain. The University of Göttingen together with the Technical University Berlin began an observation project in the Quesenbank using UAS (uncrewed aerial systems, colloq. “drones”) as sensor platforms for optical, multispectral and laserscanner-based data of 2022 and 2023.

Objectives

• Investigate UAS-based 2D images in a GIS environment to identify and map deadwood, burnt trees and ecological change markers
• Production of maps and shapefile layers containing lying and standing deadwood
• Statistical analysis of burnt area, wood volume, and damage markers based on the GIS analysis
• (optional) Analysis of deadwood volume using 3D point clouds

Requirements

• First experience in using QGIS or ArcGIS

• Interest in working with high-resolution UAS-based image data
• Both English or German writing is possible

Contact

Robert Jackisch (Robert.jackisch(at)tu-berlin.de), Birgitta Putzenlechner, (birgitta.putzenlechner(at)uni-goettingen.de), Michael Förster (michael.foerster@tu-berlin.de)

2023

Hans Jaenicke
Entwicklung von Machine Learning Prediction Modellen für die Baumartenerkennung in Niedersachsen

Okan Özsoy
GIS-basiertes Monitoring junger Gehölze und Klimaanpassungsstrategien von historischen Gärten in den Schlossparks Sanssouci und Babelsberg

Simon Hahn
Influence of landcover composition on soil moisture signal distribution derived from cosmic-ray neutron sensors - an analysis using remote sensing products at the research site Marquardt

Tom Jarling
Betroffenheit der terrestrischen Feuchtbiotope Brandenburgs von klimabedingten Änderungen des Grundwasserstands

2022

Nils Beier
Möglichkeiten und Grenzen einer Bewässerungsoptimierung von Jungbäumen in Lichtenberg

Rafael Zambrano
Characterizing urban heat and mitigation measures in Panama City

Cara Gerber
Aktueller Planungsstand in Ost- und Westbezirken Berlins & Entwicklungspotenziale

Jonathan Sommer
Detektion von Freifllächen in Fichtenforsten mithilfe einer Sentinel-2-Zeitreihe in Google Earth Engine

2020

Judith Feldhaus
Berliner Stadtbäume – Aktueller Zustand und zukünftiger Handlungsbedarf in Zeiten des Klimawandels.

2023

Rafael Zambrano
Characterizing urban heat and mitigation measures in Panama City

2022

Leonora Oels
Erfassung und Dokumentation der Massenvermehrung des Fichten-Borkenkäfers im Müritz-Nationalpark während der Jahre 2018-2020

Leon R. Schlenger
Global tree restoration – Approaches towards sustainable certification

Robert Wallace
Retrieval of the Leaf Area Index with remotely sensed RapidEye data in a beech forest within the national park Hainich, Germany

Nastasia Wolter
Can voluntary crowd-sourced data be used for vegetation mapping?

Anna Probst
Modellierung oberflächennaher Bodenfeuchte anhand drohnenbasierter TIR Daten und Analyse ihrer saisonalen räumlichen Verteilung in Marquardt.

2020

Josephine Goutrié
Urban blue space availability in German major cities: Providing a differentiated availability indicator considering different distances, water types and sizes.

Raphaela Edler
Analysis of vegetation development in Eastern Siberia by comparing declassified data from the CORONA Mission KH4B with modern imagery.

Pia Kräft
Analyse der Eignung von Oberflächenmodellen zur Erkennung von Windwurfflächen am Beispiel des Sturmes Xavier (2017) in Mecklenburg-Vorpommern.

2016

Maik Boytscheff
Wald als Regenerationsquelle für das Makrozoobenthos nach Pestizideintrag – eine GIS-basierte Landschatselementanalyse des Gewässernetzes in Norddeutschland.

2012

Robert Ahrberg
Carbon Footprint und Stadtstruktur – eine statistische Analyse der Zusammenhänge am Beispiel von Berlin

2011

Hendrikje Leutloff
Sind INSPIRE-kompatibele Geofachdaten für die Durchführung einer GIS-basierten SUP anwendbar? - Analyse zum Schutzgut Wasser auf Ebene des Flächennutzungsplans von Berlin
Bachelorarbeit Leutloff

2010

Lisa Heinsch
Application of a cellular automaton for modelling the land use in the metropolitan area of Berlin