|Paper title||Local incidence angle correction of Sentinel-1 data for time-series analysis of forest ecosystems in Google Earth Engine|
|Form of presentation||Poster|
In forest monitoring, multispectral optical satellite data have proven to be a very effective data source in combination with time-series analyses. In many cases, however, optical data have certain shortcomings, especially regarding the presence of clouds. Electromagnetic waves in the microwave spectrum can penetrate clouds, fog and light rain and are not dependent on sunlight. Since the launch of the Sentinel-1 satellite in 2014, providing freely available synthetic aperture radar (SAR) data in C-band, interest in SAR data has started to grow, and new methods began to be developed. After the launch of the second satellite, the Sentinel-1B, in 2016, a six day repeat cycle at the equator was achieved, while in temperate regions the temporal resolution can be 2-4 days thanks to orbit overlap. On the other hand, when processing a large amount of data in time-series analyses, it is necessary to use tools that can process them effectively and quickly enough, e.g., cloud-based platforms like Google Earth Engine (GEE). However, when analyzing forests over mountainous terrain, we can encounter a problem caused by the side-looking geometry of SAR sensors combined with the effects of terrain. To correct or normalize the effect of terrain, we can use, for example, the most known and most used method for this purpose, the Radiometric Terrain Correction developed by David Small. However, this method nor any other terrain correction methods were not available in GEE. Because of that, we wanted to create an alternative method for this platform. According to the findings that there is a linear relationship between local incidence angle and backscatter and that different land cover types have different relationship, we developed an algorithm called Land cover-specific local incidence angle correction (LC-SLIAC) for the GEE platform. Using the combination of CORINE Land Cover and Hansen et al.’s Global Forest Change databases, a wide range of different LIAs for a specific forest type can be generated for each scene. The algorithm was developed and tested using Sentinel-1 open access data, Shuttle Radar Topography Mission (SRTM) digital elevation model, and CORINE Land Cover and Hansen et al.’s Global Forest Change databases. The developed method was created primarily for time-series analyses of forests in mountainous areas. LC-SLIAC was tested in 16 study areas over several protected areas in Central Europe. The results after correction by LC-SLIAC showed a reduction of variance and range of backscatter values. Statistically significant reduction in variance (of more than 40%) was achieved in areas with LIA range >50° and LIA interquartile range (IQR) >12°, while in areas with low LIA range and LIA IQR, the decrease in variance was very low and statistically not significant. Six case studies with different LIA ranges were further analyzed in pre- and post-correction time series. Time-series after the correction showed a reduced fluctuation of backscatter values caused by different LIAs in each acquisition path. This reduction was statistically significant (with up to 95% reduction of variance) in areas with a difference in LIA greater than or equal to 27°. LC-SLIAC is freely available on GitHub and GEE, making the method accessible to the wide remote sensing community.
After requests from the GEE community, a new version of the algorithm was developed and uploaded to the GitHub repository, the LC-SLIAC_global, which can be used globally using the Copernicus Global Land Cover Layers, not only for countries in the European Union. Currently we are testing the LC-SLIAC algorithm in forests in tropical areas (in Vietnam) and next plans are to compare the results achieved in temporal and tropical forests, compare the achieved results using LC-SLIAC with similarly oriented methods, apply it for long-term time-series analysis of forest disturbances and subsequent recovery phases. Then to explain the reason of the short-term fluctuations of backscatter in time series – so test the influence of external and internal factors and to test radar polarimetric indices for change detection in long-term time series analyses.
Note: the original study based on the LC-SLIAC algorithm (except for the global version) was published in Remote Sensing journal (DOI: https://doi.org/10.3390/rs13091743).