Análisis de la relación entre la vegetación, temperatura y precipitación en las cuencas del Lago Conguillío y Laguna Verde utilizando el Índice de Diferencias Normalizadas de Vegetación (NDVI).
Loading...
Date
2025
Authors
Silva Olave, Javiera Belén
Journal Title
Journal ISSN
Volume Title
Publisher
Universidad de Concepción
Abstract
La Cordillera de los Andes alberga una notable biodiversidad y múltiples lagos de alto valor ecológico, los cuales presentan una elevada vulnerabilidad frente al cambio climático. Estos sistemas hídricos, frecuentemente localizados en áreas protegidas, enfrentan condiciones ambientales adversas que dificultan su estudio y monitoreo. Por ello, resulta fundamental aplicar metodologías indirectas para su análisis, como el estudio de la vegetación, estrechamente relacionada con procesos clave del balance hídrico, como la evapotranspiración. Bajo este contexto, se analizó la relación entre la cobertura vegetal y las variables climáticas de temperatura y precipitación en las cuencas del Lago Conguillío y Laguna Verde, ubicadas en el Parque Nacional Conguillío, Región de la Araucanía.
El análisis se centró en la estación de primavera (septiembre, noviembre, diciembre) y verano (diciembre, enero, febrero), entre los años 2005 y 2020, utilizando el Índice de Diferencia Normalizada de Vegetación (NDVI) como indicador de la cobertura vegetal. La metodología consistió en el procesamiento de imágenes satelitales (Landsat 4-5, Landsat 8-9 y Sentinel 2) mediante el programa de información geográfica QGIS. A partir del NDVI se definieron 3 categorías de vegetación: (1) vegetación densa, (2) vegetación moderada y (3) vegetación dispersa. Adicionalmente, se realizó el Análisis de Correlación de Pearson y Análisis de Componentes Principales (PCA) a través del software R.
Los resultados mostraron una correlación positiva fuerte y significativa (R = 0,92; p = < 2,2e-16 para Lago Conguillío y R = 0,91; p = < 2,2e-16 para Laguna Verde) entre la temperatura media y la vegetación densa, lo que sugiere que un aumento de las temperaturas estivales favorece el desarrollo de coberturas vegetales más consolidadas. Por el contrario, la precipitación evidenció una correlación negativa moderada y significativa (R = -0,56; p = 2,9e-06 en Lago Conguillío y R = -0,58; p = 9,9e-07 en Laguna Verde) con el NDVI, especialmente durante la primavera, posiblemente debido a la nubosidad persistente, sobresaturación del suelo o cobertura nival. El análisis espaciotemporal permitió identificar un incremento gradual del NDVI en ambas cuencas, específicamente en la estación de primavera, mientras en verano parece no tener variaciones importantes. Finalmente, este estudio deja en evidencia que el uso de índices derivados de imágenes satelitales sigue siendo una herramienta eficaz y ampliamente utilizada, pero debe ser complementada con técnicas de monitoreo robustas y locales.
The Andes Mountains host remarkable biodiversity and multiple lakes of high ecological value, which exhibit high vulnerability to climate change. These aquatic systems, often located in protected areas, face adverse environmental conditions that hinder their study and monitoring. Therefore, it is essential to apply indirect methodologies for their analysis, such as the study of vegetation, which is closely related to key processes of the water balance, such as evapotranspiration. Within this context, the relationship between vegetation cover and the climatic variables of temperature and precipitation was analyzed in the watersheds of Lake Conguillío and Laguna Verde, located in Conguillío National Park, Araucanía Region. The analysis focused on the spring (september, november, december) and summer (December, January, February) seasons, between 2005 and 2020, using the Normalized Difference Vegetation Index (NDVI) as an indicator of vegetation cover. The methodology consisted of processing satellite images (Landsat 4-5, Landsat 8-9, and Sentinel 2) through the geographic information software QGIS. Based on NDVI, three vegetation categories were defined: (1) dense vegetation, (2) moderate vegetation, and (3) sparse vegetation. Additionally, Pearson Correlation Analysis and Principal Component Analysis (PCA) were performed using R software. The results showed a strong and significant positive correlation (R = 0,92; p < 2,2e-16 for Lake Conguillío and R = 0,91; p < 2,2e-16 for Laguna Verde) between average temperature and dense vegetation, suggesting that an increase in summer temperatures favors the development of more consolidated vegetation cover. Conversely, precipitation showed a moderate and significant negative correlation (R = -0,56; p = 2,9e-06 in Lake Conguillío and R = -0,58; p = 9,9e-07 in Laguna Verde) with NDVI, especially during spring, possibly due to persistent cloudiness, soil oversaturation, or snow cover. The spatiotemporal analysis allowed the identification of a gradual increase in NDVI in both watersheds, with this phenomenon being more evident in spring, while in summer it appeared to show no major variations. Finally, this study highlights that the use of indices derived from satellite imagery remains an effective and widely used tool, but it must be complemented with robust and local monitoring techniques.
The Andes Mountains host remarkable biodiversity and multiple lakes of high ecological value, which exhibit high vulnerability to climate change. These aquatic systems, often located in protected areas, face adverse environmental conditions that hinder their study and monitoring. Therefore, it is essential to apply indirect methodologies for their analysis, such as the study of vegetation, which is closely related to key processes of the water balance, such as evapotranspiration. Within this context, the relationship between vegetation cover and the climatic variables of temperature and precipitation was analyzed in the watersheds of Lake Conguillío and Laguna Verde, located in Conguillío National Park, Araucanía Region. The analysis focused on the spring (september, november, december) and summer (December, January, February) seasons, between 2005 and 2020, using the Normalized Difference Vegetation Index (NDVI) as an indicator of vegetation cover. The methodology consisted of processing satellite images (Landsat 4-5, Landsat 8-9, and Sentinel 2) through the geographic information software QGIS. Based on NDVI, three vegetation categories were defined: (1) dense vegetation, (2) moderate vegetation, and (3) sparse vegetation. Additionally, Pearson Correlation Analysis and Principal Component Analysis (PCA) were performed using R software. The results showed a strong and significant positive correlation (R = 0,92; p < 2,2e-16 for Lake Conguillío and R = 0,91; p < 2,2e-16 for Laguna Verde) between average temperature and dense vegetation, suggesting that an increase in summer temperatures favors the development of more consolidated vegetation cover. Conversely, precipitation showed a moderate and significant negative correlation (R = -0,56; p = 2,9e-06 in Lake Conguillío and R = -0,58; p = 9,9e-07 in Laguna Verde) with NDVI, especially during spring, possibly due to persistent cloudiness, soil oversaturation, or snow cover. The spatiotemporal analysis allowed the identification of a gradual increase in NDVI in both watersheds, with this phenomenon being more evident in spring, while in summer it appeared to show no major variations. Finally, this study highlights that the use of indices derived from satellite imagery remains an effective and widely used tool, but it must be complemented with robust and local monitoring techniques.
Description
Tesis presentada para optar al título de Ingeniero/a Ambiental.
Keywords
Vegetación, Temperatura, Precipitación atmosférica Chile, Evapotranspiración, Cuencas hidrográficas