Publication | Closed Access
Monitoring Crop Phenology with MERIS Data | A Case Study of Winter Wheat in North China Plain
17
Citations
2
References
2009
Year
Unknown Venue
Earth ObservationCrop PhenologyPrecision AgricultureEnvironmental MonitoringEngineeringBotanyCropping SystemAgricultural EconomicsTerrestrial SensingEarth ScienceRemote Sensing ApplicationsPublic HealthCrop MonitoringMeteorologyGeographyCrop YieldAgricultureEarth Observation DataRemote Sensing TechniqueClimatologyDroughtCrop ScienceRemote SensingMeris DataOptical Remote SensingNorth China PlainRemote Sensing Sensor
Crop phonology monitoring is an important part of growth monitoring. On the other hand, the analysis on crop phenology can improve the accuracy of crop classiflcation and crop yield estimation. Crop phenology mainly has relation not only to weather variety, but also to the regional planting habit. Crop phonological stages and growth period vary in difierent areas and difierent years. With the development of remote sensing technique, the detection of crop phenology and it s mechanism using remote sensing data on regional or global scales have become popular topics in remote sensing applications. The MERIS data can provide time-serial terrestrial parameters at a several-day frequency, with which we can track the growing process of crops (take winter wheat for example) and study its variation in the growing season. The normalized difierent vegetation index (NDVI) derived from red band and near infrared band of MERIS sensor is a directly remote sensing indicator that re∞ects crop growth situation. However, due to reasons such as the in∞uence of cloud and atmospheric conditions, the residual noise in the time-series NDVI derived from MERIS will induce erroneous result in crop phenology monitoring. Thus in this study, after the time-series NDVI was computed from MERIS data, a Savitzky-Golay fllter was used to smooth out noise in NDVI time-series at pixel scale, and a time-series NDVI dataset at day frequency was produced (reconstructed). Then indicators such as the peak were extracted from the crop NDVI proflle for each pixel. After that the relation between these indicators and difierent phonological stages for winter wheat was analyzed and a model to estimate certain phonological stage for winter wheat was developed from the analysis. The model was validated with the fleld observation data collected from Fengqiu, Henan province and Yucheng, Shandong province. The validation result shows that the error in monitoring result for Heading Date and Flowering date of winter wheat is less than 3 days.
| Year | Citations | |
|---|---|---|
Page 1
Page 1