Downscaling estadístico de variables hidroclimáticas en el centro-sur de Chile: desarrollo y análisis de campos temporales mensuales y diarios en alta resolución.
Cargando...
Fecha
2025
Autores
Título de la revista
ISSN de la revista
Título del volumen
Editor
Universidad de Concepción
Resumen
A pesar de la importancia de la precipitación en el ciclo hidrológico y en diversos sectores como la agricultura, la gestión de recursos hídricos y la adaptación al cambio climático, su variabilidad física, temporal y espacial sigue representando un desafío para su simulación y proyección a escala local (Pham et al., 2021). En Chile, la interacción entre su compleja orografía, la influencia del océano Pacífico, la continentalidad y los vientos predominantes del oeste en el sur configuran un sistema climático único, lo que incrementa el desafío en la representación de esta variable (Sarricolea et al., 2017; Araya-Osses et al., 2020a; Martínez-Retureta et al., 2021).
Datos meteorológicos espaciales coherentes con la información observacional son fundamentales para varios campos científicos (Parra et al., 2004; Hijmans et al., 2005; Abatzoglou, 2013; Cannon et al., 2015b; Liu XiaoMang et al., 2017; Sun et al., 2018). Sin embargo, a pesar de la existencia de numerosos conjuntos de datos de precipitación a nivel mundial, la representación de su variabilidad espacial sigue siendo una limitante, especialmente en regiones con topografía compleja. En este contexto, los productos de precipitación en alta resolución (<1 km) son preferidos, ya que permiten capturar con mayor precisión la variabilidad ambiental que puede perderse a menor resolución.
Este estudio propone el desarrollo de una metodología integral de downscaling estadístico para estimar y analizar la variabilidad espacio-temporal de la precipitación en la zona centro-sur de Chile, entre las regiones de O’Higgins y Los Ríos. La investigación se organiza en dos en foques principales que corresponden a los productos generados: (a) Generación de un conjunto de datos de precipitación mensual en alta resolución (800 m) para el período 2000-2011. Es te producto se construye mediante la integración de múltiples fuentes de información: modelos numéricos regionales, datos satelitales, productos de reanálisis y observaciones in situ. Las correcciones sistemáticas aplicadas permiten reflejar de manera coherente la variabilidad climática de la región a escala mensual. (b) Desarrollo de un campo diario en alta resolución denominado DA-SDM (800 m), una metodología avanzada de downscaling estadístico específicamente diseñada para la estimación de eventos extremos diarios. Este enfoque utiliza información multifuente y considera factores orográficos y dinámicos para capturar adecuadamente estos eventos extremos, demostrando su capacidad para reproducirlos durante el período de mayo a agosto de 2006. Además, se destaca su potencial de aplicación en otras regiones con características similares.
Los resultados de esta tesis proporcionan dos productos complementarios: (1) un conjunto de datos de precipitación mensual en alta resolución que captura la variabilidad climática regional; y (2) un campo diario DA-SDM que permite la evaluación detallada de eventos extremos diarios. La integración de múltiples fuentes de información y la coherencia con la realidad observacional proporcionan una base sólida para estudios hidrológicos, modelación climática y planificación de recursos hídricos.
Despite the fundamental importance of precipitation in the hydrological cycle and vari ous sectors such as agriculture, water resource management, and climate change adaptation, its physical, temporal, and spatial variability remains a major challenge for accurate simulation and projection at local scales (Pham et al., 2021). In Chile, the complex interplay between orography, the influence of the Pacific Ocean, continentality, and the prevailing westerly winds in the south creates a distinctive climatic system that complicates the accurate representation of precipitation (Sarricolea et al., 2017; Araya-Osses et al., 2020a; Martínez-Retureta et al., 2021). High-quality meteorological datasets that are consistent with observational data are essen tial for a wide range of scientific applications (Parra et al., 2004; Hijmans et al., 2005; Abatzoglou, 2013; Cannon et al., 2015b; Liu XiaoMang et al., 2017; Sun et al., 2018). However, despite the availability of numerous global precipitation datasets, accurately capturing the spatial variability of precipitation remains a critical limitation, particularly in regions with complex topography. In this context, high-resolution precipitation products (<1 km) are preferred, as they provide a finer repre sentation of environmental variability that would otherwise be missed at coarser resolutions. This study presents a comprehensive statistical downscaling framework aimed at estimating and analyzing the spatio-temporal variability of precipitation across the south-central region of Chile, spanning from the O’Higgins to Los Ríos regions. The research is structured around two main objectives, each corresponding to a specific product: (a) Development of a high-resolution monthly precipitation dataset (800 m) for the period 2000–2011. This dataset is generated through the integration of multiple information sources, including regional numerical models, satellite ob servations, reanalysis products, and in situ measurements. Systematic correction procedures are applied to ensure that the dataset accurately reflects the climatic variability of the region at a monthly scale. (b) Generation of a high-resolution daily field referred to as DA-SDM (800 m). This advanced statistical downscaling methodology is specifically designed to estimate and analyze daily extreme precipitation events. By combining multi-source information and incorporating orographic and dy namic factors, DA-SDM effectively captures extreme events and demonstrates robust performance in reproducing these events during the period from May to August 2006. Additionally, this method ology offers the flexibility to be applied to other regions with similar characteristics. The outcomes of this thesis consist of two complementary products: (1) a high-resolution monthly precipitation dataset that accurately captures regional climatic variability, and (2) a spe cialized high-resolution daily field (DA-SDM) tailored for the detailed assessment of extreme pre cipitation events. The integration of diverse data sources and the consistency achieved with ob servational records provide a reliable foundation for hydrological studies, climate modeling, and water resource management.
Despite the fundamental importance of precipitation in the hydrological cycle and vari ous sectors such as agriculture, water resource management, and climate change adaptation, its physical, temporal, and spatial variability remains a major challenge for accurate simulation and projection at local scales (Pham et al., 2021). In Chile, the complex interplay between orography, the influence of the Pacific Ocean, continentality, and the prevailing westerly winds in the south creates a distinctive climatic system that complicates the accurate representation of precipitation (Sarricolea et al., 2017; Araya-Osses et al., 2020a; Martínez-Retureta et al., 2021). High-quality meteorological datasets that are consistent with observational data are essen tial for a wide range of scientific applications (Parra et al., 2004; Hijmans et al., 2005; Abatzoglou, 2013; Cannon et al., 2015b; Liu XiaoMang et al., 2017; Sun et al., 2018). However, despite the availability of numerous global precipitation datasets, accurately capturing the spatial variability of precipitation remains a critical limitation, particularly in regions with complex topography. In this context, high-resolution precipitation products (<1 km) are preferred, as they provide a finer repre sentation of environmental variability that would otherwise be missed at coarser resolutions. This study presents a comprehensive statistical downscaling framework aimed at estimating and analyzing the spatio-temporal variability of precipitation across the south-central region of Chile, spanning from the O’Higgins to Los Ríos regions. The research is structured around two main objectives, each corresponding to a specific product: (a) Development of a high-resolution monthly precipitation dataset (800 m) for the period 2000–2011. This dataset is generated through the integration of multiple information sources, including regional numerical models, satellite ob servations, reanalysis products, and in situ measurements. Systematic correction procedures are applied to ensure that the dataset accurately reflects the climatic variability of the region at a monthly scale. (b) Generation of a high-resolution daily field referred to as DA-SDM (800 m). This advanced statistical downscaling methodology is specifically designed to estimate and analyze daily extreme precipitation events. By combining multi-source information and incorporating orographic and dy namic factors, DA-SDM effectively captures extreme events and demonstrates robust performance in reproducing these events during the period from May to August 2006. Additionally, this method ology offers the flexibility to be applied to other regions with similar characteristics. The outcomes of this thesis consist of two complementary products: (1) a high-resolution monthly precipitation dataset that accurately captures regional climatic variability, and (2) a spe cialized high-resolution daily field (DA-SDM) tailored for the detailed assessment of extreme pre cipitation events. The integration of diverse data sources and the consistency achieved with ob servational records provide a reliable foundation for hydrological studies, climate modeling, and water resource management.
Descripción
Tesis presentada para optar al grado de Doctor en Ciencias Físicas
Palabras clave
Cambios climáticos, Climatología sinóptica, Hidrometeorología