High Spatial Resolution Sea Surface Climatology From Landsat Thermal Infrared Data英文电子书.pdfVIP

Remote Sensing of Environment 90 (2004) 293–307 /locate/rse High spatial resolution sea surface climatology from Landsat thermal infrared data Jeremy I. Fisher *, John F. Mustard Geological Sciences, Brown University, Box 1846, 02906, Providence, RI, USA Received 5 June 2003; received in revised form 6 January 2004; accepted 12 January 2004 Abstract High spatial and temporal resolution maps of sea surface temperature (SST) have numerous applications in coastal and estuarine systems. A climatology map, tracking SST as a function of year-day, was produced at Southern New England using 53 Landsat TM and ETM+ thermal infrared data. A recursive curve-fitting algorithm was used to fit these data and eliminate cloud contamination, resulting in an average daily temperature at every 60-m pixel. The climatology was validated against long-term in situ records that were analyzed with the same techniques. The results show, as expected, that isolated and shallow water bodies undergo more extreme temperature variation ( 2 to 25 jC) than deeper, well-connected embayments (1 to 21 jC) or the coastal ocean (4 to 18 jC). The coastal ocean is shown to lag insolation and shallow lakes by up to 44 days, with embayments showing a gradation between these extremes. Despite the subtle temperature range variation, there is rich detail in the spatial patterns which are relevant to the applied sciences of coastal and estuarine systems. The spatial pattern of the climatology reveals anomalous patterns, such as occur where anthropogenic forcing alters climatological patterns. The heat budget of Mount Hope Bay in northeas