Remote Sensing Phenology



USGS Remote Sensing Phenology Data Quality Note9/30/2013Data quality issues have been observed in the USGS (MODIS and AVHRR) phenological metrics and these issues can usually be tracked to certain causes, such as: Excessive noise in the source eMODIS or AVHRR NDVI time series data due to atmospheric contamination that we were unable to remove by our data smoothing algorithmTime-varying sensor degradation in the MODIS instrument (e.g. ADDIN EN.CITE <EndNote><Cite><Author>Wang</Author><Year>2012</Year><RecNum>499</RecNum><DisplayText>(Wang et al., 2012)</DisplayText><record><rec-number>499</rec-number><foreign-keys><key app="EN" db-id="s5tdvzzfw0p0x8ev5r85exvoz9s2rp2f55pa">499</key></foreign-keys><ref-type name="Journal Article">17</ref-type><contributors><authors><author>Wang, D.</author><author>Morton, D. C.</author><author>Masek, J.</author><author>Wu, A.</author><author>Nagol, J.</author><author>Xiong, X.</author><author>Levy, R.</author><author>Vermote, E.</author><author>Wolfe, R. E.</author></authors></contributors><titles><title>Impact of sensor degradation on the MODIS NDVI time series</title><secondary-title>Remote Sensing of Environment</secondary-title></titles><periodical><full-title>Remote Sensing of Environment</full-title></periodical><pages>55-61</pages><volume>119</volume><number>0</number><keywords><keyword>Sensor degradation</keyword><keyword>MODIS</keyword><keyword>NDVI</keyword><keyword>Time series</keyword></keywords><dates><year>2012</year></dates><isbn>0034-4257</isbn><urls><related-urls><url>;(Wang et al., 2012)Failure in the phenological metrics software to properly extract a feature, such as the trend change identified as the Start of Season Time (SOST) or End of Season Time (EOST). This generally occurs when the variability in the source time-series NDVI is too low to trigger identification. These are known issues that can contribute to an inaccurate characterization of one or more phenological feature expressed in the annual metrics or lead to image artifacts. The Remote Sensing Phenology team at USGS EROS is committed to improving the quality of our landscape-level phenological data. To address data quality issues that can be tracked to time-varying MODIS sensor degradation, we anticipate reprocessing the entire historical time-series of 2001-2012 eMODIS 250 m metrics when Collection 6 (C6) MODIS Products listed in Table 1 are made available. These data are currently scheduled for release late in 2013. Planned changes in sensor calibration in the context of the C6 reprocessing should produce time series that have higher quality and are more consistent (i.e., less evidence of sensor degradation trends) than the C5 MODIS data. Table 1. MODIS data inputs used to generate the eMODIS Conterminous U.S. NDVI time-series dataProductDescriptionCollection VersionAlgorithm Version*MOD09MODIS/Terra Surface Reflectance 5-Min L2 Swath 0055.3.29MOD03MODIS/Terra Geolocation Fields 5-Min L1A Swath 1km0055.0.44MOD35_L2MODIS/Terra Cloud Mask and Spectral Test Results 5-Min L2 Swath 250m and 1km0055.3.1*as of July 26, 2011For further identification of the specific locations of these issues in the phenology data, please see the Known Issues links found at . References Cited ADDIN EN.REFLIST Wang, D., Morton, D. C., Masek, J., Wu, A., Nagol, J., Xiong, X., . . . Wolfe, R. E. (2012). Impact of sensor degradation on the MODIS NDVI time series. Remote Sensing of Environment, 119(0), 55-61. doi: 10.1016/j.rse.2011.12.001 ................
................

In order to avoid copyright disputes, this page is only a partial summary.

Google Online Preview   Download