Validation of AVHRR and TMI-derived sea surface temperature in the northern South China Sea

Chunhua Qiu,Dongxiao Wang,Hiroshi Kawamura,Lei Guan,Huiling Qin
DOI: https://doi.org/10.1016/j.csr.2009.10.009
IF: 2.629
2009-01-01
Continental Shelf Research
Abstract:Satellite-derived SSTs are validated in the northern South China Sea (NSCS) using in situ SSTs from the drifting buoys and well-calibrated sensors installed on Research/Vessel(R/V) Shiyan 3. The satellite SSTs are Advanced Very High Resolution Radiometer (AVHRR) daytime SST, AVHRR nighttime SST, Tropical rainfall Measuring Mission Microwave Imager (TMI) daytime SST and TMI nighttime SST. Availability of satellite SST, which is the ratio that the number of available satellite SST to the total ocean pixels in NSCS is calculated; annual average SST availabilities of AVHRR daytime SST, AVHRR nighttime SST, TMI daytime SST and TMI nighttime SST are 68.42%, 69.99%, 56.57% and 52.80%, respectively. Though the TMI SST availability is nearly constant throughout the year, the variations of the AVHRR SST availability are much larger because of seasonal variations of cloud cover in NSCS. Validation of the satellite-derived SSTs shows that bias±standard deviation (STD) of AVHRR SST is −0.43±0.76 and −0.33±0.79°C for daytime and nighttime, respectively, and bias±STD of TMI SSTs is 0.07±1.11 and 0.00±0.97°C for daytime and nighttime, respectively. It is clear that AVHRR SSTs have significant regional biases of about −0.4°C against the drifting buoy SSTs. Differences between satellite-derived−in situ SSTs are investigated in terms of the diurnal SST cycle. When satellite-derived wind speeds decrease down below 6m/s, the satellite SSTs become higher than the corresponding in situ SSTs, which means that the SST difference (satellite SST−Buoy SST) is positive. This wind-speed dependence of the SST difference is consistent with the previous results, which have mentioned that low wind speed coupled with clear sky conditions (high surface solar radiation) enhance the diurnal SST amplitude and the bulk-skin temperature difference.
What problem does this paper attempt to address?