Career

High-Resolution Soil Moisture Mapping Over Arctic Permafrost Regions Using an Unmanned Aerial Vehicle (UAV)-Based Thermal Imagery
 

MSc level

Description: The rapid increase in air temperatures (2-4 times as fast as the global average) in high-latitude regions causes enhanced permafrost thaw, and related shifts in uptake and release of carbon. This process might potentially increase the atmospheric concentrations of greenhouse gases (GHGs) such as CO2 and CH4, which might in turn cause an increase in the rate of warming (positive feedback). The permafrost carbon processes and the impact of the environmental conditions need to be investigated to improve our understanding to what extent and how fast this increase will occur. One of the most important environmental factors that governs the permafrost carbon cycle is soil moisture, which affects e.g. microbial activities and organic matter decomposition rates. Therefore, assimilating soil moisture in any upscaling model will improve the estimation accuracy of the obtained carbon budgets.
In the project outlined here, soil moisture will be estimated based on thermal imageries that will be collected by an unmanned aerial vehicle (UAV), which is capable of surveying large areas with high resolution. The produced soil moisture maps will complement the carbon flux measurements of various platforms (i.e., manual chambers, a UAV platform, and an eddy tower) over the Canadian Arctic to provide a better estimation of the carbon budget.
Responsibilities: 

• Conducting a literature survey on the UAV-based thermal imaging and soil moisture estimation methodologies, including machine learning approaches
• Processing and analyzing the UAV-based thermal images to estimate the soil moisture content
• Supporting carbon flux upscaling effort

Useful skills: 

• Basic knowledge of a scripting language like R or python
• Good communication skills (oral and written) in English
• Basic understanding about statistical analysis

Timeframe: Field work in the Canadian Arctic is planned for late June to July 2024. Participation in this campaign would be beneficiary.
Contact: Abdullah Bolek (abolek@bgc-jena.mpg.de)

Frequency and intensity of extreme drought events in state-of-the-art climate model projections

BSc / MSc level

Description: The CMIP (Coupled Model Intercomparision Project) is a global initiative to  better understand past, present and future climate change impacts on ecosystems and vegetation . Therefore, the project coordinates simulations of most available state-of-the-art Earth System Models. The output of these models are usually a collection of global climate variables (e.g. temperature, precipitation) at high resolution time-scales, but also vegetation and ecosystem states (e.g. primary productivity, soil water content). Task of the student is to identify regions of extreme drought stress in these simulations using (1) well established indicators of drought stress, and also (2) more modern machine learning based techniques. The goal of the thesis is to identify agreement and differences between the Earth System Models and to relate these to relevant processes of the model eventually feeding back to the CMIP community.
Useful skills: 

• Profound knowledge of a script programming language such as Python or R
• No fear of working with datasets that can have more than 100GB in size
• Basic understanding about the relevant processes and variables of the global carbon cycle

Contact: Phillip Papastefanou (papa@bgc-jena.mpg.de)

Identifying key plant hydraulic traits of European species using the terrestrial ecosystem model QUINCY

MSc level

Description: Terrestrial ecosystem models (TBMs) such are QUINCY are used to investigate impacts of current and future climate projections on forests, corps and vegetation in general. Plant hydraulics, i.e  more detailed physical representations of the conducting system of plants have recently been brought into focus of TBMs as they are crucial when understanding impacts of extreme events such as droughts. Such a new plant hydraulic architecture has also been introduced to QUINCY and its capability of reproducing drought induced vegetational responds have been demonstrated. However, the influence of the new physicall parameters have not yet been tested and calibrated for the most important plant/tree species in Europe or Globally. The overall objective of this thesis is to improve our undestanding of drought impacts on vegetation by improving the performance of the QUINCY model. Thereby, the student should (1) test the newly implemented plant hydraulic architecture and (2) perform a sensitivity analysis to esimate parameter ranges and estimates for the most important European tree species.

Useful skills: 

• Knowledge of a script programming language such as Python or R to perform sensititivy or related analysis
• Experiences in sensitivity analysis based methods such as latin hypercube sampling or sobol indices
• Basic understanding about the physical processes of plants, tree and ecology in general

Contact: Phillip Papastefanou (papa@bgc-jena.mpg.de)

Stochastic modeling of tundra lakes in permafrost landscapes

Description: The novel stochastic model with which we generate the random Arctic landscapes provides very promising results. However, the statistics of the lake size distribution must be improved. This project is about the stochastic modeling of the Arctic landscape and numerical experiments on the boundary layer in terms of large eddy simulations. 
Useful skills: 

• Solid foundations in mathematics and stochastics
• Theoretical atmospheric physics and geophysical fluid dynamics
• High performance computing

Contact: Mark Schlutow (mschlutow@bgc-jena.mpg.de)

Understanding the large methane peak observed at ATTO after the 2015/2016 El Nino

BSc / MSc level

Description: In 2016, just after a strong positive phase of ENSO (El Nino), which caused strong soil moisture anomalies in the Amazon region, there was a large and sustained methane peak at the 4m level at the ATTO site. The 4-m level is the measurement height closest to the soil but the Upland forest is generally a sink of methane, not a source. In this project the student will investigate what drove such methane source and try to understand the processes looking into soil moisture and temperature, the full profile of CH₄ and CO₂ mole fractions, and other environmental variables.
Useful skills: 

• knowledge of a programming language like R or python 
• Basic knowledge of the carbon cycle and methane production processes
• knowledge on environmental physics

Contact: Santiago Botía (sbotia@bgc-jena.mpg.de)

Exploring the effect of tree mortality on the net ecosystem exchange using LPJ-GUESS

MSc level

Description: Plot based studies suggest that the carbon sink in Amazon old-growth forests is declining. Such decline seems to be driven by an increasing tree mortality that counteracts the productivity in these forests. In this project the student will analyze the output of the biosphere model LPJ-GUESS, which was modified to simulate the mortality trend observed in the old-growth forest plots. The focus is to understand how the mortality trend propagates into the net ecosystem exchange flux and potentially perform atmospheric transport simulations to study if we can eventually observe a signal of the increasing mortality in the atmosphere. 
Useful skills: 

• knowledge of a programming language like R or python 
• Some knowledge on linux and bash scripting
• Basic knowledge of the carbon cycle and ecology
• knowledge on environmental physics

Contact: Phillip Papastefanou (papa@bgc-jena.mpg.de) and Santiago Botía (sbotia@bgc-jena.mpg.de)