Topographic effects on levelling comparisons: case studies of Saint-Paul-de-Fenouillet (Eastern Pyrenees) and Arudy (Western Pyrenees), France

Differences between repeated levelling surveys measure vertical movements. We show that these comparisons are correlated with the topography of the me asured lines. This correlation constitutes an important additional source o f error in levelling. We analyse levelling lines in two regions in the Fren ch Pyrenees where earthquakes occurred recently (Saint-Paul-de-Fenouillet a nd Arudy) and correct the levelling comparisons using the correlation betwe en the topographic slope and the measured tilt. This correction reduces the vertical signature by a factor of 2. This demonstrates that it is absolute ly necessary to take into account the topographic effect in analysing and i nterpreting levelling comparisons. ((C) 1999 Academie des sciences/Editions scientifiques et medicales Elsevier SAS.).