Authors: Pu Huang*, Department of Computing Sciences, Texas A&M University-Corpus Christi, Yuxia Huang, Faculty of Department of Computing Sciences, Texas A&M University-Corpus Christi
Topics: Remote Sensing, Applied Geography, Coastal and Marine
Keywords: Self-organizing map, Total suspended matter, MODIS, Yellow River estauary
Session Type: Poster
Start / End Time: 9:55 AM / 11:35 AM
Room: Lincoln 2, Marriott, Exhibition Level
Presentation File: No File Uploaded
Total suspended matter (TSM) is playing an important role in water quality assessment and resource management. The Moderate-resolution Imaging Spectroradiometer (MODIS) 250 m imagery has been proven useful to map the concentration of TSM. In this study, a long-term reflectance image dataset during the period from 2000 to 2018 was constructed to map the concentration of TSM in Yellow River estuary, China, based on MODIS 250 m imagery. An unsupervised learning method, self-organizing map (SOM) was applied to this long-term dataset to identify underlying TSM patterns. Several TSM patterns were identified from the long-term dataset using a two-dimensional SOM of size 3 x 4 grids. Frequencies of the different TSM patterns were calculated and analyzed, and characteristics of the TSM patterns and factors affect the variability of the TSM patterns were discussed. This study also demonstrated a general approach to apply SOM to MODIS data to identify TSM patterns in coastal waters.