SP 09: Landscape scale land cover analysis and geodata management

Managing land use conflicts considering carbon flux optimization, securing biodiversity, food production, and economic stability requires profound and up-to-date knowledge on land use. Development paths have to be identified to gather improved knowledge for likely future land use scenarios. Remote sensing analyses and land use modelling provide such input consistently, in a spatio-temporally explicit way, and across scales. The aim of this subproject is to analyse landscape scale land use- and land-cover change to support decision-making for an optimized land management. We develop and apply a landscape-wide analysis approach integrating remote sensing and spatial modelling techniques to gain knowledge on how to mitigate existing and prevent future land use conflicts. The project focuses on:

• Development of remote sensing based analysis schemes to derive land use at high
  resolution (e.g. Landsat data) with regional coverage over the last 25 years
• Adaptation of machine learning and time series algorithms to cope with large datasets
• Development and application of remote sensing based indicators for assessing landscape
  patterns and its links with carbon sequestration potential at landscape level
• Spatially explicit modelling on landscape level to identify drivers and hot spots of change
  at the landscape scale
• Spatially explicit scenario-building of land use types according to different regional to
  sub-continental storylines

In addition a spatial data infrastructure for the whole carbiocial project including
data management and web-based technologies for distributed data access in Germany and
Brazil is provided.

Please see also the Poster of SP 09 presented at the Kick-Off Workshop in Cuiabá, Brazil, 6-9 July 2011

Project Members
Prof. Dr. Patrick Hostert
Prof. Dr. Tobia Lakes
Dr. Pedro J. Leitão
PhD Student Hannes Müller
PhD Student Florian Gollnow
Geomatics Lab, Geography Department, Humboldt University Berlin