Haiti: damage assessment using pictometry
EEFIT team members, Keiko Saito (CAR), team leader Edmund Booth and Gopal Madabhushi (Department of Engineering, Cambridge University)
Following the Haiti earthquake of 12 January, 2010, CAR director Keiko Saito took part in a reconnaissance mission by the British Earthquake Engineering Field Investigation Team (EEFIT) to Port-au-Prince. The Mw 7.0 earthquake caused devastation to this already impoverished nation, the poorest in the western hemisphere. The earthquake happened at 16:53 local time, when people were still at work and in school. More than 200,000 people are estimated to have been killed by the earthquake, a staggering number amounting to approximately 2% of the total population of Haiti.
The international community mobilised to coordinate the relief efforts in collaboration with the incapacitated national government. Given the scale of the disaster, the Post-Disaster Needs Assessment (PDNA) process was initiated, where the World Bank had the mandate to estimate the reconstruction cost for Haiti.
Cambridge Architectural Research took part in the PDNA, by taking on the task of assessing the damage distribution of the buildings, using Pictometry (high resolution oblique aerial photographs) data for the first time. Pictometry captures oblique views, making the façade of the buildings visible on the image, ideal for damage assessment. This data became the basis for estimating the reconstruction cost.
The EEFIT team surveyed 137 buildings to validate the damage assessment carried out using remotely sensed data. The distribution of the damage levels derived using both datasets were compared, which revealed that 50% of the completely collapsed buildings, which are the most obvious damage types visible in the images, were captured using Pictometry. This is the first time that damage assessment using images has been validated against field survey data. CAR is continuing to investigate ways to improve the damage assessment methodology.
Download:Damage assessment using Pictometry.] (PDF 772 Kb)