Land / Biosphere

Mapping of the functional diversity of forests

Productivity and stability of forest ecosystems strongly depend on the functional diversity of plant communities. Forests with higher functional diversity are generally more productive and stable over long timescales than less diverse forests. Diverse plant communities show increased resource use efficiency and utilization, enhanced ecosystem productivity and stability and can better cope with changing environmental conditions – an insurance effect of biodiversity. They are also less vulnerable to diseases, insect attacks, fire and storms. With remote sensing, you can measure and map functional diversity of forests at different scales – from individual trees to whole communities.

  Spatial composition of the physiological traits leaf chlorophyll, carotenoids and water content. The colour composite shows the relative abundance of the three traits.
Spatial composition of the physiological traits leaf chlorophyll, carotenoids and water content. The colour composite shows the relative abundance of the three traits. (Image: UZH/RSL/F. Schneider)


Classifying types of cultivation sequences 

Frequent satellite observations of Switzerland not only allow the generation of very current products but also the monitoring of change over time. Typical changes over a year also reveal useful information on the usage of areas, as is the case for agriculture. Based on satellite based time-series regions which are likely used for agricultural were identified and afterwards classified into easily identifiable categories. The categories are related to vegetation growing seasons, the patterns of which can be retrieved from the time-series. The resulting categories only allow limited interpretation but illustrate what differentiations between field usage are possible.

Categories of cultivation sequences (e.g. one or two phases of crop growth within) based on variations in the seasonal vegetation dynamics of crops.
Categories of cultivation sequences (e.g. one or two phases of crop growth within) based on variations in the seasonal vegetation dynamics of crops. (Image: UZH/NPOC/D. Fawcett)


Generating cloud-free mosaics of satellite images

Optical data has the drawback of cloud-cover obscuring surface information and thus many acquisitions can only partly be used for the mentioned products. To cover a certain region, many acquisitions must be composited together (i.e. combined to a mosaic) and ere therefore often a patchwork of most recent cloud-free observations. In the example of the Landsat-8 mosaic presented here, the oldest commonly date back 2 months for regions that are often cloudy and not as frequently captured.

Cloud free mosaic based on Landsat-8 and Sentinel-2 images for late spring 2018.
Cloud free mosaic based on Landsat-8 and Sentinel-2 images for late spring 2018. (Image: UZH/NPOC)


Mapping of soil properties

An increasing demand for full spatio-temporal coverage of soil information drives the growing use of soil spectroscopy. With multi-temporal imaging spectroscopy data, it is possible to increase the total mapping area of bare soils in a heterogeneous agricultural landscape. The final multi-temporal composite image contained more than double the amount of bare soil pixels as compared to a singular acquisition. As a result, an improved spatial representation of soil parameters such as percentage of clay in the topsoil layer can be retrieved from the composite data.

Barest pixel soil composite processing chain
Barest pixel soil composite processing chain (S. Diek et al., Remote Sensing 9 (12), 2017, doi:10.3390/rs9121245