Harmonization of Landsat and Sentinel 2 for Crop Monitoring in Drought Prone Areas: Case Studies of Ninh Thuan (Vietnam) and Bekaa (Lebanon)
Minh D. Nguyen,Oscar B. Villanueva,Duong D. Bui,Phong T. Nguyen,Lars Ribbe
DOI: https://doi.org/10.20944/preprints201910.0275.v1
2019-10-24
Abstract:Proper satellite-based crop monitoring applications at the farm-level often require near-daily imagery at medium to high spatial resolution. The synthesizing of ongoing satellite missions by ESA (Sentinel 2) and NASA (Landsat7/8) provides this unprecedented opportunity at a global scale; nonetheless, this is rarely implemented because these procedures are data demanding and computationally intensive. This study developed a complete stream processing in the Google Earth Engine cloud platform to generate harmonized surface reflectance images of Landsat7,8 and Sentinel 2 missions. The harmonized images were generated for two agriculture schemes in Bekaa (Lebanon) and Ninh Thuan (Vietnam) during the period 2018-2019. We evaluated the performance of several pre-processing steps needed for the harmonization including image co-registration, brdf correction, topographic correction, and band adjustment. This study found that the miss-registration between Landsat 8 and Sentinel 2 images, varied from 10 meters in Ninh Thuan, Vietnam to 32 meters in Bekaa, Lebanon, and if not treated, posed a great impact on the quality of the harmonized dataset. Analysis of a pair overlapped L8-S2 images over the Bekaa region showed that after the harmonization, all band-to-band spatial correlations were greatly improved from (0.57, 0.64, 0.67, 0.75, 0.76, 0.75, 0.79) to (0.87, 0.91, 0.92, 0.94, 0.97, 0.97, 0.96) in bands (blue, green, red, nir,swir1,swir2, ndvi) respectively. We demonstrated that dense observation of the harmonized dataset can be very helpful for characterizing cropland in highly dynamic areas. We detected unimodal, bimodal and trimodal shapes in the temporal NDVI patterns (likely cycles of paddy rice) in Ninh Thuan province only during the year 2018. We fitted the temporal signatures of the NDVI time series using harmonic (Fourier) analysis. Derived phase (angle from the starting point to the cycle's peak) and amplitude (the cycle's height) were combined with max-NDVI to generate an R-G-B image. This image highlighted croplands as colored pixels (high phase and amplitude) and other types of land as grey/dark pixels (low phase/amplitude). Generated harmonized datasets that contain surface reflectance images (bands blue, green, red, nir, swir1, swir2, and ndvi at 30 meters) over the two studied sites are provided for public usage and testing.