Land cover characterization for Colombia at 25m resolution using radar RS data. Source: Marcela Quiñones at


MSc thesis subject: Land cover characterization for Colombian climate- and global change studies: comparison of 5 minutes to 50m resolution land cover datasets for Colombia

Within the Meteorology and Air Quality group we apply a meteorological and air quality modelling system called WRF (Weather & Research Forecasting system). This model uses default a land cover and land use dataset according to the USGS (US Geological Survey) but recently, in a study with a focus on Colombia, we have introduced the use of IMAGE present-day and future land cover and land use and have also access to a 50m radar-based land cover data for Colombia. We propose an MSc thesis study to evaluate these different land cover datasets for Colombia.

In a recent MSc study using WRF to assess the impact of land cover and land use (LUC) changes on Colombian meteorology and air quality, we have introduced use of LUC data according to the IMAGE modelling system. WRF uses default a LUC dataset according to the USGS also distinguishing quite different LUC classes compared to IMAGE. Main motivation to introduce the use of IMAGE LUC data is that this integrated assessment model system provides, besides data on present-day LUC, also future scenarios on LUC. However, the resolution of these data (5 minutes) is coarse, especially compared to another land cover dataset available for Colombia based on radar remote sensing observations at a resolution of 50m. The MSc thesis study with a focus on Colombia indicated that differences between the present-day USGS and IMAGE LUC data already results in simulated changes in meteorology that are larger compared to the simulated changes considering the future and present-day LUC. It expresses that at the moment we are still limited in assessing the impact of future LUC changes on climate and global change due to these uncertainties associated with LUC characterization.

We invite students with a background in remote-sensing (and GIS) to conduct an MSc thesis study with the focus on evaluating these different LUC datasets for Colombia. The study should also result in a system that translates the IMAGE LUC classification into the WRF LUC classification that results in the most optimal representation of present-day LUC according to the radar-based high resolution LUC dataset.


  • To compare three different land cover datasets on their characterization of land cover in Colombia at different spatial resolutions ranging from 5 minutes to 50m resolution.
  • To transform the land cover characterization according to IMAGE into a classification that provides the best representation of LUC for Colombia for application in the WRF modelling system.


  • Feddema, J., Oleson, K., Bonan, G., Mearns, L., Washington, W., Meehl, G., and Nychka, D. (2005). A comparison of a GCM response to historical anthropogenic land cover change and model sensitivity to uncertainty in present-day land cover representations. Climate Dynamics, pages 581-609.


  • Background in RS and GIS acquired in the MSc MGI courses

Theme(s): Sensing & measuring, Integrated Land Monitoring