Climate Trend Analysis and AI-Based Forecasting for Nordic Capitals
Moreira, Jhonny (2026)
Julkaisun pysyvä osoite on
https://urn.fi/URN:NBN:fi:amk-202603104110
https://urn.fi/URN:NBN:fi:amk-202603104110
Tiivistelmä
The objective of this thesis was to design and implement an interactive web-based dashboard for viewing and evaluating historical climate data from five Nordic capitals: Helsinki, Copenhagen, Oslo, Reykjavik and Stockholm. The project was motivated by the importance of climate change in the Nordic region, where temperatures have been rising at a faster rate than the global average, as well as a lack of accessible tools for investigating and comparing long-term climate records across these cities.
The project was done as applied development research. Climate data was obtained from the European Climate Assessment and Dataset (ECA&D), which includes daily temperature records from 1873 for Copenhagen to 1951 for Helsinki. The dashboard was built in Python with Streamlit, Plotly, and Prophet, and the application was deployed on the Streamlit Community Cloud.
The historical analysis confirmed a consistent warming trend across all cities, with Helsinki and Stockholm experiencing the biggest mean annual temperature increase between 1951-1980 and 2000-2025 at +1.78 °C. The Prophet model predicts a consistent seasonal pattern for 2026-2031, with Copenhagen remaining the warmest city and Reykjavik the coldest. Future work could extend the tool’s capabilities by including new climate variables and strengthening the forecasting module with models that consider external climate influences.
The project was done as applied development research. Climate data was obtained from the European Climate Assessment and Dataset (ECA&D), which includes daily temperature records from 1873 for Copenhagen to 1951 for Helsinki. The dashboard was built in Python with Streamlit, Plotly, and Prophet, and the application was deployed on the Streamlit Community Cloud.
The historical analysis confirmed a consistent warming trend across all cities, with Helsinki and Stockholm experiencing the biggest mean annual temperature increase between 1951-1980 and 2000-2025 at +1.78 °C. The Prophet model predicts a consistent seasonal pattern for 2026-2031, with Copenhagen remaining the warmest city and Reykjavik the coldest. Future work could extend the tool’s capabilities by including new climate variables and strengthening the forecasting module with models that consider external climate influences.