Integrating remote sensing data from multiple optical sensors for ecological and crop condition monitoring

Abstract

Ecological and crop condition monitoring requires high temporal and spatial resolution remote sensing data. However remote sensing instruments trade spatial resolution for swath width and it’s difficult to acquire remotely sensed data with both high spatial resolution and frequent coverage. A synthesized approach fusing multiple types of remote sensing imagery provides a feasible and economical solution. In this paper, we demonstrate an operational data fusion framework based on Spatial and Temporal Adaptive Reflectance Fusion Model (STARFM) for integrating existing MODIS data products (daily, 500m) and freely available Landsat data (16-day, 30m). Phenological metrics are extracted from the fused Landsat and MODIS data. Our case study focuses on an agricultural region in central Iowa. Initial results show that the detailed spatial and temporal variability of the landscapes can be identified from the fused remote sensing data. The derived phenology metrics show distinct features for crops and forest at the field scales and can be explained by the USDA’s reports on the crop progress. The data fusion and time-series analysis approaches provide a feasible solution to for ecological and crop condition monitoring at the field scales.