Landslide inventories are essential for the elaboration of susceptibility and hazard maps using data-driven methods, as well as for the validation and evaluation of the performance of heuristic or physically based methods. The best landslide susceptibility map corresponds to a detailed landslide inventory. At the national level in Colombia, there are two databases that record landslide occurrence: SIMMA run by the Colombian Geological Service and Desinventar run by the Seismological Observatory of Southwestern Colombia (OSSO). In this work, a landslide inventory for the northern Colombian Andes was elaborated using free tools such as Google Earth online, Google Earth Pro and QGIS, which have available satellite images from optical and microwave sensors with sub-meter resolution since the 1980s. For each event, landslide crown and propagation zone were identified, as well as the time interval in which the landslide could first be identified in the available imagery. A total of 610 landslides were identified from 2000 to 2020, with areas between 255 m2 and 1’036,685 m2.  For the same period 492 landslides were recorded for SIMMA and 1,236 for DesInventar. The results of this study shows that remote sensing could be used for validating landslide databases such as Desinventar, in which a huge number of landslide reports have been registered through a telephone call to the emergency number 123 or to the offices of the Municipal Councils for Disaster Risk Management. Many of these events do not really correspond to landslides and are not subject to a field visit, due in many cases to the large number of reports received, especially during the rainy season. In terms of localization, remote sensors reduce the levels of uncertainty associated with the data in databases such as Desinventar or SIMMA. The results indicate the importance of the use of remote sensors, many of which, like those used in this study, are open access, in order to identify a greater number of events and reduce uncertainty in spatial location. 0

  • Natalia Andrea Pino Serna, Edier Aristizabal, Anika Braun, Martin Tzschoppe
  • Poster