Modelos para la estimación del rendimiento de la caña de azúcar en Costa Rica con datos de campo e índices de vegetación
[EN] Remote sensing offers important inputs for sugarcane yield estimation, since its temporal and spatial approaches allow monitoring the phenological cycle of the crop. The objective of this research was to apply an operational method for the estimation of sugarcane yield and sugar content through...
| Autores: | , , |
|---|---|
| Tipo de recurso: | artículo |
| Fecha de publicación: | 2023 |
| País: | España |
| Institución: | Universitat Politècnica de València (UPV) |
| Repositorio: | RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia |
| Idioma: | español |
| OAI Identifier: | oai:riunet.upv.es:10251/191683 |
| Acceso en línea: | https://riunet.upv.es/handle/10251/191683 |
| Access Level: | acceso abierto |
| Palabra clave: | Sugarcane Vegetation indexes Linear regression Sentinel-2 Landsat-8 Caña de azúcar Índices de vegetación Regresiones lineales |
| Sumario: | [EN] Remote sensing offers important inputs for sugarcane yield estimation, since its temporal and spatial approaches allow monitoring the phenological cycle of the crop. The objective of this research was to apply an operational method for the estimation of sugarcane yield and sugar content through the combination of field variables with vegetation indices, calculated with the satellite sensors on board Sentinel-2 and Landsat-8 in a cooperative from Costa Rica. In addition, historical harvest data and start months of phenological cycle were used to estimate sugarcane yield and sugar content per ton using multiple linear regressions. The integration of historical data and Simple Ratio (SR) vegetation index, calculated in different steps of the phenological cycle (in the months of September, December and January), allowed us to obtain an estimation model of sugarcane yield (tons of sugarcane per hectare) with a regression coefficient (R2) of 0.64 and a RMSE of 8.0 tons/ha. While for sugar content (kilograms of refined sugar per ton) we obtained a R2 of 0.59 integrating historical variables and the vegetation indexes SR and Green Normalized Difference Vegetation Index (GNDVI); in this case the RMSE was 4.9 kg/tons. Ultimately, this operational method of yield estimation provides tools for decision making before, during and after the harvest stage. |
|---|