Merging Satellite Images for High Spatial and Temporal Resolution of Lake Güija

Abstract

This study aims to evaluate and compare satellite image fusion techniques to improve the spatial and temporal resolution of data for high-precision local studies [1][2]. The availability of detailed, small-scale information is essential for the monitoring and management of natural resources, especially in areas with productive activities or potential environmental impacts, such as Lake Güija. This lake, located on the border between Guatemala and El Salvador, flows into the Lempa River, the main source of drinking water and agriculture in the region [3].

For this purpose, images from the Landsat and MODIS satellites were used, selected for their complementary observation capabilities at different spatial and temporal resolutions [4]. Two fusion approaches were applied: the spatial Pansharpening method, which combines high-resolution panchromatic data with multispectral imagery to optimize spatial detail, and the spatio-temporal STARFM (Spatio-Temporal Adaptive Reflectance Fusion Model), which integrates Landsat and MODIS datasets to obtain data sets with improved spatial and temporal resolution [5] [6] [7].

The results show that both techniques are viable for generating useful information in local studies, with the Pansharpening-based methodology with MODIS input being the most appropriate due to its balance between quality and implementation feasibility. Lake Güija is presented as a case study that illustrates the potential of these methodologies for the continuous and accurate monitoring of aquatic ecosystems, thus contributing to the availability of reliable data for decision-making.