Detecting Methane Anomalies from Satellite Data in the Arctic

The escalating climate crisis is fueled by greenhouse gas emissions. Right after carbon dioxide the second most powerful anthropogenic greenhouse gas is methane. Relevant and reliable information about the sources and sinks of methane is needed to mitgate climate change and its effects. This is particularly true in the Arctic, where the rapidly warming climate poses unprecedented environmental threats, such as melting permafrost, which could become a new methane source. As these changes continue to unfold, an increasing amount of satellite data is becoming available to observe them. However, the Arctic also has many features that restrict good quality satellite observation, which makes it a difficult area to study. In this thesis, methane anomalies are detected by comparing smaller and larger spatial area averages. Significant differences between these values could indicate abnormal behavior and even environmental changes, such as new methane sources, in the research area. For this anomaly detection TROPOMI WFMD v1.8 dataset in the Arctic region for the summer and spring seasons of 2018--2023 is used. The main focus of the thesis is to introduce a method for creating an anomaly set and to evaluate if the TROPOMI WFMD v1.8 dataset has enough good quality data with an even distribution for atmospheric research in the high latitudes. The aim of the thesis is to provide a dataset of anomalies that a domain expert could use as a starting point for research. The method consists of three data processing steps: satellite data binning, anomaly detection and identifying important anomalies. The dataset constructed with the newly proposed method showed some areas with consistent anomalous behavior. These anomalies could be used to find new anthropogenic and natural sources of methane, for example in the melting permafrost regions. However, the thesis concluded, that even with the large data volume, satellite data from the region is unevenly distributed and thus prone to biases. The biases can be reduced through methods introduced in the thesis, but further studies for data gap filling in the high latitudes could improve methane emission estimates made from satellite products.

Grantor dc:publisher

Helsingin yliopisto

Year dc:date.issued

2024

Author dc:creator

• Nevala, Elma Elviira

• emissions
• methane
• greenhouse gases
• arctic region
• anomaly detection
• remote sensing
• satellite data
• outlier detection

Statement dc:rights

• CC BY 4.0

Language dc:language.iso

eng

Handle dc:identifier.uri

http://hdl.handle.net/10138/589265

OAI identifier oai:identifier

oai:helda.helsinki.fi:10138/589265