3D-TAIGER: Multi-Source 3D Geoinformation Extraction for Improved Management of Forest and Natural Hazards – Collaboration between TAIwan and GERmany
Project-related news can be found in our GIScience News Blog.
The main scientific objective of our exchange and collaborative research is the development of new methods for the automatic extraction of 3D geoinformation related to forests and natural hazards in Taiwan using multi-source remote sensing data. The project 3D-TAIGER is an international research collaboration and networking activity between the Institute of Geography, GIScience, LiDAR Research Group (LRG) of the Heidelberg University (HU) and the Department of Geomatics and Department of Earth Sciences of the National Cheng Kung University (NCKU) in Tainan/Taiwan. It is funded jointly by the DAAD (Germany) and MOST (Taiwan).
Research and Collaboration Activities
The cooperation between the teams from NCKU and HU is beneficial for both parties: The HU team has a wealth of experience with ALS data processing, from ALS intensity correction, point cloud processing to geographic information extraction and applications; while the NCKU team has in-depth understanding in remote sensing of the local tropical forest and natural hazards, and has access to abundant remote sensing data sets, including a very high density full-waveform ALS data and hyperspectral data. Within this two year project, we plan to conduct several joint workshops, and field experiments and data acquisition. For the HU team, the tropical forest and natural hazards in Taiwan provide new scientific challenges for existing methodologies and can help to enhance the robustness and transferability for a global setting. Here, the rich experience of the NCKU team in remote sensing of complex geographic settings of natural hazards in tropical forests will help the HU to revise existing methods and to create entirely new geoinformation extraction concepts by knowledge exchange. Furthermore, the HU team will particularly benefit from the NCKU experience of combining ALS and passive hyperspectral remote sensing.
This MOST/DAAD project will enhance the competences in remote sensing based 3D geoinformation extraction of the two groups in terms of both i) methodological and ii) geographic knowledge exchange.
One specific outcome of this project will be the extension of the International LiDAR Vegetation Signature Database – LVISA (http://www.uni-heidelberg.de/lvisa) hosted at HU, which contains tree species LiDAR information. LVISA will be extended with ALS reference signatures covering Taiwan’s most prominent tree species. Further, LVISA’s Web analysis tools will be extended to support carbon contribution estimation by NCKU.
A specific study site has already been selected for which the joint method development will take place between HU and NCKU teams. The proposed Taiwan study site is in Tsengwen reservoir of Southern Taiwan (see Figure below). Remote sensing data, including a high density ALS full-waveform dataset and a hyperspectral dataset, are readily available.