The third factor is called windy factor. This factor is significant positively loaded to
maximum wind speed for mainland regions and also loaded to average wind speed for the
northern and central regions. Time series of windy-factor score show peaks corresponding
with that of warm factor for the mainland region, but they are not significantly different
from annual mean. However, high variations of windy-factor scores were found among
individual mainland stations. This suggests that local forcing(s) could be more influencing
the wind characteristics. For the southern east coast, the windy factor is positively loaded to
average wind speed and negatively loaded to maximum temperature. This indicates that
higher average wind speed here is commonly found with lower maximum temperature here
and could imply strong circulation driven by land-sea heat exchanges in December and
January, corresponding to the attack of northeast winter monsoon (see Figure 3d). This
windy factor accounts for 21.4% to 25.1% of the total variation
December is not dry and the anomaly in DJF was high (-351 to 795 mm). This high anomaly
correlated to strong ENSO and IOD events occurring in previous two month (October to
December, OND), as suggested by high canonical correlation coefficient (r = 0.31 to 0.48)
over the South (see Figure 4 DJF). From Table 2, positive Indian Ocean Dipole and negative
Southern Oscillation Index (SOI), indicating El Nino, occurring in OND was likely found
with negative DJF rainfall anomaly over the southern east coast. Also, La Nino event in OND
was with more DJF rainfall. Moreover from Table 2, both ENSO and IOD strengths in OND
was equally associated with Thailand rainfall; whereas ENSO was dominant from January to
May and IOD was singly dominant in summer monsoon month (June to August, JJA).