Evaluación de la estructura del bosque mediante sensores remotos y validación en campo, en el Pacífico Norte de Costa Rica.

This research focuses on evaluating forest structure by integrating field data and remote sensing, using satellite images from Landsat, Sentinel 1, and Sentinel 2, to develop a methodology, algorithms, and maps. The approach is based on the use of multivariable models to provide a comprehensive characterization of the forest, with an emphasis on basal area, aboveground biomass, and species diversity. The study will begin with the creation of a comprehensive database from previous study plots and the establishment of new plots. This database will include information on diameter at breast height, total height, and tree species for trees with a diameter greater than 10 cm, as well as the number of observed vertical strata. Additionally, basal area, biomass, and species diversity will be calculated for each plot. The organization of this data will allow for an accurate characterization of the plo
Next, a multivariate analysis will be conducted using variables derived from field data and remote sensing data. Variables such as surface temperature calculated from Landsat images, elevation provided by a Digital Elevation Model (DEM) from NASA, a surface roughness index obtained from cross-polarization data (VV/VH) from Sentinel 1, and vegetation and texture indices calculated with Sentinel 2 will be used. Multivariate techniques will assist in the ordering, classification, and statistical analysis of the variables to determine their relationships and identify the most influential ones. The model will be constructed using 80% of the sample data and validated with the remaining 20%. This model will integrate field and remote variables to predict basal area, aboveground biomass, and species diversity, as well as the combination of these variables that will represent forest structure. Once built, the model will be adjusted based on the r
Finally, the validated model will be applied to the satellite images to analyze the spatial variability of basal area, aboveground biomass, species diversity, and forest structure in the study area. This analysis will allow the identification of forest zones in different stages of development. Thematic maps and descriptive statistics will be generated to provide a detailed view of the forest development status.
In summary, the research combines advanced remote sensing techniques and multivariate analysis with field data to provide a comprehensive assessment of forest structure. The results will offer valuable information for forest management and conservation, facilitating a deeper understanding of forest conditions and supporting decision-making in natural resource management.

Desarrollar una nueva clasificación del bosque en función de su estructura, con sensores remotos.

Gestión para la conservación y restauración de ecosistemas naturales.