Academic works

2024

1. nature

   

   An annual land cover dataset for the Baltic Sea Region with crop types and peat bogs at 30 m from 2000 to 2022

   2024

   Abs HTML PDF

   We present detailed annual land cover maps for the Baltic Sea region, spanning more than two decades (2000-2022). The maps provide information on eighteen land cover (LC) classes, including eight general LC types, eight major crop types and grassland, and two peat bog-related classes. Our maps represent the first homogenized annual dataset for the region and address gaps in current land use and land cover products, such as a lack of detail on crop sequences and peat bog exploitation. To create the maps, we used annual multi-temporal remote sensing data combined with a data encoding structure and deep learning classification. We obtained the training data from publicly available open datasets. The maps were validated using independent field survey data from the Land Use/Cover Area Frame Survey (LUCAS) and expert annotations from high-resolution imagery. The quantitative and qualitative results of the maps provide a reliable data source for monitoring agricultural transformations, peat bog exploitation, and restoration activities in the Baltic Sea region and its surrounding countries.

2023

1. EGU

   

   Crop type mapping in Central and South Asia using Sentinel-1 and Sentinel-2 remote sensing data

   Christoph Raab, and Viet Duc Nguyen

   2023

   Abs HTML Poster

   Crop type information derived from satellite remote sensing are of pivotal importance for quantifying crop growth and health status. However, such spatial information are not readily available for countries in Central and South Asia, where smallholder farmers play a dominant role in agricultural practice, and food security. In this study, we provide insights into crop type mapping for three study sites in the region: 1) Panfilov District in Kazakhstan, 2) Jaloliddin Balkhi District in Tajikistan, and 3) Multan District in Pakistan. A collection of Sentinel-2 and Sentinel-1 satellite data was used along with the random forest classification algorithm. To train and validate the classification model, field data were collected between May and October 2022 in each of the study areas. Our main objective was to evaluate the performance of a combined Sentinel-2 and Sentinel-1 mapping approach in comparison to a single source result. In addition, this contribution will provide insights into the performance with regard to crop type mapping accuracy of different temporal data aggregation intervals. Preliminary results indicate a small increase in overall accuracy for a combined Sentinel-2 and Sentinel-1 mapping approach. However, Sentinel-2 data might be sufficient for reliable crop type mapping, in case cloud coverage is not a constraint. Future studies might consider evaluating the potential benefit of using a full Sentinel-1 data set without temporal aggregation for mapping crop types.