Bibliographic citations
Nuñez, J., (2022). Desarrollo de un modelo de evaluación y monitoreo de áreas degradadas de pastizal con sistemas de información espacial [Universidad Nacional Agraria La Molina]. https://hdl.handle.net/20.500.12996/5519
Nuñez, J., Desarrollo de un modelo de evaluación y monitoreo de áreas degradadas de pastizal con sistemas de información espacial []. PE: Universidad Nacional Agraria La Molina; 2022. https://hdl.handle.net/20.500.12996/5519
@phdthesis{renati/244565,
title = "Desarrollo de un modelo de evaluación y monitoreo de áreas degradadas de pastizal con sistemas de información espacial",
author = "Nuñez Delgado, Jimny Yoel",
publisher = "Universidad Nacional Agraria La Molina",
year = "2022"
}
An experiment was conducted with the objective of generating a spatial evaluation and monitoring model of degraded grassland areas, using indicators of attributes of the conservation status of pajonal and puna grass ecosystems that can be detected with remote sensors. The evaluation was carried out in the districts of Simón Bolívar, Canchayllo and Santa Ana, in the central highlands of Peru. Data collection and sample collection were carried out in the years 2015, 2016 and 2017. 120 transects were installed within a rigid network of 250 x 250 m resolution following the coordinates of the satellite images of the MODIS sensor. Soil (SoilGrid, Aster) and vegetation (LandSat) databases were used. Seven study variables were identified: maximum NDVI, elevation, erosion/deposition ratio, soil organic matter, bulk density, plant cover, and aerial biomass. The field work included measurements of organic matter (samples taken to the laboratory), spectral reflectance (EM50 Decagon Devices field sensor), and aerial biomass (cut and separation). For the analysis of the information, the Random Forest model was used to classify grasslands as very poor, poor, regular, good and excellent. The model chose the soil organic matter variable as the most important, followed by aerial biomass, which had a global adjustment of 60% at the time of calibration, categorizing grasslands as very poor (0%), poor (34.12%), regular (53.85%), good (12.01%) and excellent (0.013%). Finally, the degradation model was compared with the RESTREND spatial model, confirming that the grasslands of Santa Ana (Huancavelica) have a greater negative trend (15.27%) since they have no response to precipitation. It is concluded that soil organic matter and plant biomass can be spatialized and monitored with global models and remote sensors, respectively, to predict the state of conservation of grasslands and puna grass.
This item is licensed under a Creative Commons License