The diurnal amplitude of sea surface temperature (SST) at 1-m depth is estimated with a simple empirical model in the region of 80¿E--160¿W, 60¿S--60¿N covering the western Pacific Ocean using 4 years of satellite-derived wind speed, solar radiation, and latent heat flux data. The advantage of this indirect method is that the diurnal amplitude of SST (ΔSST) can be obtained even in cloudy areas. ΔSST is large in the tropics through the year. Summer ΔSST in the middle and high latitudes of the Northern Hemisphere, especially around Japan, is larger than that of the Southern Hemisphere. However, ΔSST in the Kuroshio and its extension regions is smaller compared with that in its surrounding ones because of higher wind speed. In the Bay of Bengal, the South China Sea, and the Timor Sea, ΔSST becomes largest in spring, when the surface wind becomes weak in the transition of the monsoon. The period of about 35--50 days is clearly seen in the temporal variation of ΔSST in the tropics. This reflects the Madden-Julian oscillation. Furthermore, there are some differences in ΔSST between El Ni¿o and La Ni¿a periods. The authors then investigate how clouds affect the detection of ΔSST. If sampling is limited to clear areas only, the areas of strong solar radiation and weak wind are selectively picked up, and the probability that larger ΔSST is sampled increases. The mean of the ΔSST sampled from clear areas only is greater by 0.2--0.3 K compared with the actual mean.