Skip to main content
Fig. 6 | Geoenvironmental Disasters

Fig. 6

From: Landslide Susceptibility Mapping of Urban Areas: Logistic Regression and Sensitivity Analysis applied to Quito, Ecuador

Fig. 6

This shows the p-values of the K-S test (two-sample) applied to 2000 landslide susceptibility datasets resulting from the univariate susceptibility analysis of the LOGIT model by weights discretization. Legend: The alpha value to determine the significance level was 0.05, which is marked with the dashed red line. The values (dots) above this level (precipitations, population, and road density) represent the non-significant samples from 13 simulations, whereas the remaining 1987 are considered to be at significance levels. The building footprint factor dots (ochre color) cover most of the dots representing these remaining simulations, which are very close to a p-value of zero. Lines with dots represent factors: 1geo = lithology (electric blue), 2cov = land use/vegetation coverage (orange), 3sei = seismicity (light gray), 4pre = precipitations (yellow), 5sta = soil stability (light blue), 6slo = slope (green), 7pop = population (dark blue), 8roa = road density (brown), 9bui = floor area (dark gray) and, 10gro = building footprint area (ochre). X-axis shows the multipliers of each factor’s coefficient and Y-axis shows the p-values

Back to article page