18
T.A. Cinco et al. / Atmospheric Research 145–146 (2014) 12–26
1.5
Anomaly (base yr 61-90)
1998, 1.0
Temperature Anomlay (0C)
1.0
5 years running mean
0.5
Linear (5 years running mean )
0.0
y = 0.017x - 0.3679
-0.5
1950
1952
1954
1956
1958
1960
1962
1964
1966
1968
1970
1972
1974
1976
1978
1980
1982
1984
1986
1988
1990
1992
1994
1996
1998
2000
2002
2004
2006
2008
2010
-1.0
Year
Fig. 4. Observed mean annual minimum temperature anomalies in the Philippines during the 1951–2010 period (compared with 1961–1990 normal values).
the 1997 El Niño event peaked during the period October–
December (OND) and thereafter began to decay, finishing in
the period March–May (MAM) of 1998. In the Philippines, an
El Niño (La Niña) event is normally associated with drier
(wetter) than normal conditions with the most severe events
resulting in droughts (floods). However, more recent studies
have shown that there is sub-national variability within
this generalisation with the reverse (i.e. wetter conditions)
occurring during some events (Lyon et al., 2006; Villafuerte
et al., 2014).
The large 1998 anomaly could also be associated with a
global maximum in the same year which is considered one of
the warmest in the last century based on instrumental records
of global surface temperatures from three major climatic
monitoring organisations around the world. Zonal data for the
equator to 24°N from NASA's GISS Surface Temperature
Analysis (GISSTemp), clearly shows 1998 and 2010 (which
is the second warmest year since 1951 in the Philippines
according to the Tmean indicator) as the hottest years since 1880
although the base period for this is 1951–1980, not 1961–1990
as in this study. This is also reflected in the UK Met Office's
HadCRUT3 dataset which shows a clear peak in 1998 for global
temperatures and a downturn in the early 1990's (Brohan et al.,
2005). This provides an indicator that this aspect of the
observed data in the Philippines is consistent with global data
(IPCC, 2007, 2013a).
4.2. Extreme daily events
Analysis of trends in extreme daily maximum and minimum temperatures (hot-days index and cold-nights index,
respectively) show that there are a statistically significant
increasing number of hot days and decreasing number of cold
nights. Fig. 5 shows the trends in the frequency of days with
minimum temperature below the 1st percentile (cold nights)
and Fig. 6 the trends in frequency of days with maximum
temperature above the 99th percentile (hot days). In each
figure an increase is represented by a (+) sign, a significant
increase with (▲); decreases are shown using a (−) symbol
and significant decreases with a (▼) symbol.
Table 2 shows in detail the calculated values used to
determine the direction and significance of the trends for the
figures above for each of the 30 synoptic stations at which
the trend was observed. Those stations which showed a
statistically significant trend are highlighted in grey. This
demonstrates that there is a statistically significant increase
in the occurrence of hot days and decrease in cold nights
across much of the Philippines during the period 1951–2010.
In fact none of the stations shows a statistically significant
increase in cold nights throughout the Philippines with
only five of the thirty stations showing any observed
decrease (statistically insignificant). Those stations which
show significant decreases in cold nights are dispersed
throughout the Philippines on all major island groups
although the station at General Santos, the southernmost
station considered here, shows a particularly large statistically significant decrease.
There is greater variation between stations when we
consider the hot days index with five of the 30 stations
exhibiting a statistically significant decrease in the occurrence
of hot days. Stations in the northern part of the Philippines
(Baguio, Cabanatuan, Infanta and Laoag) show mainly an
increase in hot days, four of them at a statistically significant
level. Zamboanga station shows strong, statistically significant