The Research Of Yellow River Ice Flood Detection Model Based On EOS MODIS Data

Yellow River is the second longest river in China, but in all ages, its annual ice flood threatened people's safety and property in the Yellow River basin. Yellow River's ice flood is due to its especial geographic condition that its flowing direction is from low latitude to high latitude. In the freezing up and melting period, if there appears seriously ice barrier or ice dam in the reach, the water level will be increased which easily lead to floodplain and crevasse. Such disasters threaten a safety and wealth of people, who live along the river. Therefore ice detection of Yellow River is very important. Every year the Yellow River workgroup will use mass of manpower and material resource in ice flood prevention.Remote sensing technology is large-scale and continuous, which is very appropriate for large range observation, exp Yellow River. Using the remote sensing data, the commission can detect those positions where are iced over or not ice up and take necessary precaution measures to avoid ice flood. Yellow River Conservancy Commission using remote sensing data in the Yellow River ice detection in 2002 for the first time, and MODIS data is applied widely as the primary data in this monitoring work.MODIS (the Moderate Resolution Imaging Spectroradiometer) on Earth Observation Satellites is on the polar orbiting Earth Observation System (EOS) for day and night global observations. For the high spatial resolution, high spectrum resolution and high temporal resolution of MODIS, the timely monitoring becomes possible. The MODIS snow and ice cover algorithm is based on the reflectance characteristic of snow. In this paper Yellow River's ice is identified with four criteria, Normalized difference Snow Index (NDSI), the reflectance of a near infrared test for detecting water, and the reflectance test for cloud. A model based on these criteria is presented in this paper. By comparing to the timely in situ data from the Yellow River hydrometnc station everyday, the model was developed and the result was validated. The achievements in this work are described in the paper and the results show that it's more efficient when the method was used in the extraction of the information from the satellite data. Satellite data is expected to predict the ice flood in the future. Finally, we made software based on this model and via the feedback of the user, it was developed and perfected.