Multi-satellite data fusion to monitor soybean growth and deforestation over Brazil

Jian Peng, University of Illinois at Urbana-Champaign

Usage Details

Jian Peng, Yaping Cai, Kaiyu Guan, Bin Peng, Bin Peng, Chongya Jiang, Wang Zhou, Tianyu Sun, Yi Yang, Yizhi Huang, Kejie Zhao, Jingwen Zhang, Ziqi Qin, Ziyi Li, Zewei Ma, Sheng Wang, Qu Zhou, Maxwell Jong, Trevor Wong, Melissa Chen, Han Qiu, Xiaocui Wu, Zhenrui Yue

There  is  tradeoff  between  spatial  and  temporal  resolutions  for  any  single  satellite  mission.  Multi-sensor data fusion is a promising approach to achieve both high spatial and temporal resolutions. Here we will use Blue Waters to scale up the MODIS-Landsat fusion project to the 12 States of the U.S. Midwest through our newly developed SaTellite DAta IntegRation (STAIR) algorithm.

MODIS has a daily revisit frequency but relatively lower spatial resolution (500m for surface reflectance product MCD43A4).  Landsat  has  a  30-m  spatial  resolution  but  lower  revisit  frequency  (16  days).  The  expected product  from  this  fusion  project  is  a  30-m resolution,  daily,  cloud-free  and  spatio-temporally  continuous surface  reflectance  dataset  for  the  midwestern  United  States,  which  will  enable  novel  agricultural applications in both academia research and industry, such as estimations of surface downward radiation, evapotranspiration, crop phenology stage, crop yield, and crop type classification, at unprecedentedly high resolutions in both space and time.

The  computational  demand  of  producing  such  high  resolution  and  daily  satellite  images  through fusion is  huge  and the Blue Waters facility offers us  the  best solution for large processing element demands, high-frequency  I/O,  and  output  post-processing  and  visualization.