作者: Bertrand Saulquin , Francis Gohin , René Garrello
DOI: 10.1109/TGRS.2010.2052813
关键词: Ocean color 、 SeaWiFS 、 Meteorology 、 Variogram 、 Kriging 、 Environmental science 、 European union 、 Remote sensing 、 Moderate-resolution imaging spectroradiometer 、 Covariance 、 Imaging spectrometer
摘要: In this paper, we define the method that is used to merge high-resolution multisensor chlorophyll-a (chl-a) data on Ireland-Biscay-Iberia Regional Ocean Observing System area from 1998 present at a resolution of 1.1 km. The based geostatistics and known as kriging. merged variable daily anomaly chl-a, with being defined difference between image mean historical field for considered day. For each day, continuous generated using kriging method, then added obtain cloudless chl- a. initial satellite chl-a set in merging procedure derived level-2 water leaving radiances three ocean color sensors: Sea-Viewing Wide Field View Sensor Orbview platform, Moderate Resolution Imaging Spectroradiometer Aqua Medium Spectrometer Instrument ENVISAT platform. concentration obtained specific algorithm developed by Ifremer, OC5 product. Before merging, satellite-derived product has been compared situ validated matchup set. After validation against data, intercomparisons sets have performed. As variability depends, region, season distance shore, local space-time semivariograms calculated estimate spatiotemporal dependence or covariance chlorophyll anomalies. semivariograms, estimation kriged anomaly, are their nuggets (noise), spatial temporal range (maximum nonnull anomalies), sill variance). semivariogram approximated locally regular grid. sills deduced square (the climatological reference) observed classical proportionality effect mean, variance distribution, parameters semivariograms. Compared original product, analysis shows complete coverage differences statistically equivalent those offers number applications environmental monitoring such eutrophication risk required Water Framework Directive European Union.