TY - JOUR

T1 - Sea surface temperature spatio-temporal variability in the Azores using a new technique to remove invalid pixels

AU - Lafon, Virginie

AU - Martins, Ana

AU - Bashmachnikov, Igor

AU - Melo-Rodrigues, Margarida

AU - Figueiredo, Miguel

PY - 2004/5/3

Y1 - 2004/5/3

N2 - Since 4th April 2001, Sea Surface Temperature (SST) of Azorean waters is obtained from 1.1 km resolution NOAA-12, -14, and -16 imagery recorded at the "HAZO" HRTP mid-Atlantic satellite receiving station. One year of data is processed to investigate AVHRR-derived SST distributions and associated dominant space and time scales. Daily SST is calculated with the MultiChannel Sea Surface Temperature algorithm. Comparison with in situ historical data demonstrates that satellite SST ranges include low value pixels that may be attributed to undetected clouds. Images show that remnant clouds are associated to zones of strong cloud coverage. However, contamination of pixels in contact with clouds is found to be limited. Therefore, to remove erroneous values, rather than systematically erode pixels around clouds, images are filtered inputting to each image threshold values equal to the mean more or less 4 times the standard deviation of 8-day SST histograms. Nighttime SST averages show a distinct seasonal cycle for the Azores region. Increased surface heating and decreased horizontal gradients are found during summertime. Eight-day zonal and meridional SST averages show typical latitudinal and longitudinal gradients, providing some insight into the physical mechanisms responsible for surface temperature variability in the region. These data are relevant to improve stock assessment and fishery management studies in the Azores.

AB - Since 4th April 2001, Sea Surface Temperature (SST) of Azorean waters is obtained from 1.1 km resolution NOAA-12, -14, and -16 imagery recorded at the "HAZO" HRTP mid-Atlantic satellite receiving station. One year of data is processed to investigate AVHRR-derived SST distributions and associated dominant space and time scales. Daily SST is calculated with the MultiChannel Sea Surface Temperature algorithm. Comparison with in situ historical data demonstrates that satellite SST ranges include low value pixels that may be attributed to undetected clouds. Images show that remnant clouds are associated to zones of strong cloud coverage. However, contamination of pixels in contact with clouds is found to be limited. Therefore, to remove erroneous values, rather than systematically erode pixels around clouds, images are filtered inputting to each image threshold values equal to the mean more or less 4 times the standard deviation of 8-day SST histograms. Nighttime SST averages show a distinct seasonal cycle for the Azores region. Increased surface heating and decreased horizontal gradients are found during summertime. Eight-day zonal and meridional SST averages show typical latitudinal and longitudinal gradients, providing some insight into the physical mechanisms responsible for surface temperature variability in the region. These data are relevant to improve stock assessment and fishery management studies in the Azores.

KW - Azores

KW - Image filtering

KW - Spatial variability

KW - SST

KW - Temporal variability

UR - http://www.scopus.com/inward/record.url?scp=1942517420&partnerID=8YFLogxK

U2 - 10.1117/12.514249

DO - 10.1117/12.514249

M3 - Conference article

AN - SCOPUS:1942517420

VL - 5233

SP - 89

EP - 97

JO - Proceedings of SPIE - The International Society for Optical Engineering

JF - Proceedings of SPIE - The International Society for Optical Engineering

SN - 0277-786X

T2 - Remote Sensing of the Ocean and Sea Ice 2003

Y2 - 9 September 2003 through 12 September 2003

ER -