Evaluación del impacto de la densidad de flujo acumulado en la zonificación de susceptibilidad a movimientos en masa en la microcuenca La Presidenta (Medellín)
Fecha
2025
Título de la revista
ISSN de la revista
Título del volumen
Editor
Universidad de Manizales
Referencia bibliográfica
Resumen
La microcuenca La Presidenta, ubicada en el municipio de Medellín, Antioquia, enfrenta un alto riesgo de movimientos en masa debido a su topografía abrupta, su régimen hídrico y un proceso de urbanización acelerado. Aunque en Colombia se han desarrollado metodologías para la zonificación de amenaza, la versión vigente propuesta por el Servicio Geológico Colombiano (2017) no contempla la densidad de flujo acumulado como variable condicionante, lo que podría limitar la capacidad del modelo para capturar dinámicas locales relevantes en la ocurrencia de estos eventos. Esta investigación evalúa el impacto de incorporar la densidad de flujo acumulado en el análisis de susceptibilidad a movimientos en masa en la microcuenca La Presidenta. Para ello, se compilaron registros históricos y secundarios de eventos tipo deslizamiento, los cuales sirvieron como base para aplicar un modelo de zonificación mediante el método de Pesos de Evidencia (WoFE). Posteriormente, se comparó el desempeño de la metodología estándar con una versión modificada que incluye la nueva variable, utilizando análisis estadísticos (curva de éxito) y validación espacial. Los resultados revelan que la densidad de flujo acumulado presenta el mayor peso estadístico dentro del modelo, destacando su alto valor explicativo para representar procesos de escorrentía que inciden en la inestabilidad del terreno. Aunque la mejora porcentual en la curva de éxito fue leve (de 70.49 % a 70.5 %), la inclusión de esta variable permitió ajustes significativos en zonas urbanizadas cercanas a corrientes de agua, donde la susceptibilidad pasó de baja o media a alta, refinando la representación espacial del riesgo. Este efecto se observó principalmente en sectores con infraestructura crítica o dinámicas de drenaje complejas.
No obstante, se identificaron limitaciones en áreas sin historial de eventos, como la zona media-alta de la cuenca, donde la densidad de flujo acumulado no mejoró la predicción. Esto sugiere una dependencia importante de la calidad y cobertura del inventario de procesos morfodinámicos para que esta variable exprese su potencial predictivo.
En conjunto, la investigación aporta una metodología robusta y replicable que combina enfoques estadísticos con criterios heurísticos, y demuestra que la densidad de flujo acumulado puede enriquecer la zonificación de susceptibilidad, siempre que se cuente con información base suficiente. Los hallazgos contribuyen al fortalecimiento de los procesos de planificación territorial y gestión del riesgo en territorios con características similares.
The La Presidenta micro-watershed, located in the municipality of Medellín, Antioquia, faces a high risk of mass movements due to its steep topography, hydrological regime, and accelerated urbanization. Although Colombia has developed methodologies for hazard zoning, the current version proposed by the Colombian Geological Survey (2017) does not consider flow accumulation density as a conditioning variable, which may limit the model's ability to capture relevant local dynamics in the occurrence of these events. This research evaluates the impact of incorporating flow accumulation density into the landslide susceptibility analysis in the La Presidenta micro-watershed. To this end, historical and secondary records of landslide-type events were compiled and used as the basis to apply a zoning model using the Weight of Evidence (WoFE) method. The performance of the standard methodology was then compared to a modified version that includes the new variable, using statistical analyses (success curve) and spatial validation. The results show that flow accumulation density has the highest statistical weight within the model, highlighting its strong explanatory value for representing runoff processes that influence terrain instability. Although the percentage improvement in the success curve was small (from 70.49% to 70.5%), the inclusion of this variable led to significant adjustments in urbanized areas near watercourses, where susceptibility shifted from low or medium to high, refining the spatial representation of risk. This effect was particularly notable in areas with critical infrastructure or complex drainage dynamics. However, limitations were identified in areas without a history of events, such as the mid-upper zone of the watershed, where flow accumulation density did not improve prediction. This suggests a significant dependency on the quality and coverage of the landslide inventory for this variable to fully express its predictive potential. Overall, this study contributes a robust and replicable methodology that combines statistical and heuristic approaches and demonstrates that flow accumulation density can enhance susceptibility zoning, provided that sufficient baseline information is available. The findings support improved territorial planning and risk management in areas with similar characteristics.
The La Presidenta micro-watershed, located in the municipality of Medellín, Antioquia, faces a high risk of mass movements due to its steep topography, hydrological regime, and accelerated urbanization. Although Colombia has developed methodologies for hazard zoning, the current version proposed by the Colombian Geological Survey (2017) does not consider flow accumulation density as a conditioning variable, which may limit the model's ability to capture relevant local dynamics in the occurrence of these events. This research evaluates the impact of incorporating flow accumulation density into the landslide susceptibility analysis in the La Presidenta micro-watershed. To this end, historical and secondary records of landslide-type events were compiled and used as the basis to apply a zoning model using the Weight of Evidence (WoFE) method. The performance of the standard methodology was then compared to a modified version that includes the new variable, using statistical analyses (success curve) and spatial validation. The results show that flow accumulation density has the highest statistical weight within the model, highlighting its strong explanatory value for representing runoff processes that influence terrain instability. Although the percentage improvement in the success curve was small (from 70.49% to 70.5%), the inclusion of this variable led to significant adjustments in urbanized areas near watercourses, where susceptibility shifted from low or medium to high, refining the spatial representation of risk. This effect was particularly notable in areas with critical infrastructure or complex drainage dynamics. However, limitations were identified in areas without a history of events, such as the mid-upper zone of the watershed, where flow accumulation density did not improve prediction. This suggests a significant dependency on the quality and coverage of the landslide inventory for this variable to fully express its predictive potential. Overall, this study contributes a robust and replicable methodology that combines statistical and heuristic approaches and demonstrates that flow accumulation density can enhance susceptibility zoning, provided that sufficient baseline information is available. The findings support improved territorial planning and risk management in areas with similar characteristics.
Descripción
Palabras clave
Sistema de Información Geográfica (SIG), Movimientos en masa, Cuencas Hidrográficas
