conditions is not always easy. The strong drought of 2007 described above, happened unexpectedly in a very strong LN year (Yumul et al., 2010), which even exacerbated the devastation of the event. Lyon et al. (2006) analyzed this connection between rainfall and ENSO by looking at rainfall data from 40 observing stations across the Philippines and showed that the seasonal rainfall response to ENSO reverses between boreal summer (July–September) and fall (October–December) for both El Niño and La Niña. On a seasonal basis, Villafuerte II et al. (2014b) found in another publication that ENSO is correlated to the precipitation in OND (October to December) in the way that El Niño leads to dryer and La Niña to wetter conditions. In contrast, the season JAS is more influenced by the monsoonal activity, in particular in the western part of the country. The trends in extreme heat events, e.g. in hot days and warm nights, have been analyzed by Manton et al. (2001), who showed significant increases and further a decrease in cool days and cold nights in the period 1961 to 1998. Projected changes of the EWEs To analyze the future changes in drought occurrence, Dai (2012) used global models, but could not find clear significant results for SEA. In contrast to that, Taylor et al. (2012) found a consistent increase in drought risk in a global assessment, which he indicates with the Palmer Drought Index (PDSI). Sillmann et al. (2013) showed that heavy precipitation events in SEA will increase in magnitude and frequency in the future. The share of heavy precipitation on the total amount of precipitation could increase to more than 50% in RCP8.5. They also found signs for a stronger drought risk as they showed that the maximum number of consecutive dry days is increasing. Looking at the superregional level of SEA, the strongest increases for frequency and intensity of temperature extremes are found to be in Indonesia and southern Philippines by the World Bank (2013). Sillmann et al. (2013) found similar results and further argued that SEA is one of the two regions in the world that face strong increases in heat extremes even under a low-emissions scenario. In the RCP8.5 scenario, today’s warm spells (above the 90th percentile of the base time period) would occur at most times of the year (~300 days) and today’s warm nights (above the 90th percentile of the base time period) would occur in 95% of all nights in the future (Sillmann et al., 2013) For the projected changes in TCs, one has to look separately for the frequency and the intensity of the storm events. As the World Bank (2013a) stated, there are no studies dealing with projections for a temperature change of only 2°C – so they all analyze future projections of TCs for a 4°c warmer world (RCP8.5). Atmospheric model results show an overall decrease in frequency in the area of the Philippines – with some exceptions (Sugi et al., 2009; Knutson et al., 2010; Held and Zhao, 2011; Murakami et al., 2012). Sugi et al. (2009) used a mesh AGCM (Atmospheric General Circulation Model) and showed also robust decreases in global TC frequency due to climate change. However, they stated that the regional characteristics can be different as they depend strongly on the SST projections used. Other working groups (Caron and Jones, 2007; Emanual et al., 2008) analyzed the likelihood of a TC to develop – which is called cyclogenesis and can be used as an indicator for frequency (World Bank, 2013a) – and found increasing trends. In contradiction, Zhao and Held (2011) proved that the 13