Time Series of Quad-Pol C-Band Synthetic Aperture Radar for the Forecasting of Crop Biophysical Variables of Barley Fields Using Statistical Techniques

This study aims to fit and predict crop biophysical variables using a series of SAR images by conducting a factorial experiment and estimating time series models through a combination of forecasts. The research focused on two barley plots grown under rainfed conditions in Spain during the 2015 growi...

Descripción completa

Detalles Bibliográficos
Autores: Sipols, Ana E., Valcarce Diñeiro, Ruben, Santos Martín, María Teresa, Sánchez, Nilda, Simón de Blas, Clara
Tipo de recurso: artículo
Fecha de publicación:2022
País:España
Institución:Universidad Rey Juan Carlos
Repositorio:BURJC-Digital. Repositorio Institucional de la Universidad Rey Juan Carlos
OAI Identifier:oai:burjcdigital.urjc.es:10115/41230
Acceso en línea:https://hdl.handle.net/10115/41230
Access Level:acceso abierto
Palabra clave:RADARSAT-2
polarimetric SAR
biophysical variables
time series
cointegration
Descripción
Sumario:This study aims to fit and predict crop biophysical variables using a series of SAR images by conducting a factorial experiment and estimating time series models through a combination of forecasts. The research focused on two barley plots grown under rainfed conditions in Spain during the 2015 growing season (February to June). The dataset included nine field measurements of agronomic parameters, 20 RADARSAT-2 images, and daily weather records. Ten polarimetric observables were extracted and integrated to derive six key agronomic and monitoring variables, including height, biomass, vegetation cover fraction, leaf area index, water content, and soil moisture. Statistical methods, such as double smoothing, ARIMAX, and robust regression, were employed to model these field variables. The model equations indicated a positive influence of meteorological factors and a significant temporal component in crop development, reflecting natural growth patterns. After combining various models, the results showed improved forecasting efficiency and highlighted the impact of several weather variables. The presence of a cointegration relationship between data series from the same crop across different fields enables the prediction of crop outcomes in similar fields without the need for re-modelling.