Correlation between monthly climate variables and mass flowering
Below you can find scatter plots of monthly weather variables and flowering in the three different regions which show strongest relationships. I chose climate variables in the months that occur prior to flowering to answer my research question if any of them may be cues for flowering. In graph 10 you can see a correlation between February precipitation and November and December SOI which had the highest correlation with flowering in Borneo. (I dropped July maximum temperature from this analysis as flowering usually starts in February or March in Borneo and even though July maximum temperature shows a high correlation, it cannot be a cue for flowering.) The graph suggests that a particularly strong drought (low precipitation values) in February and El Nino (low SOI values) in November and December may trigger flowering to occur in Borneo.
Graph 10: Scatter plot of climate and flowering correlations in Borneo
In graph 11 you can see a correlation between January and February Precipitation and flowering and April cloud cover and flowering in Malaysia. In Malaysia flowering is known to occur anywhere from March to June. Both January & February precipitation and April cloud cover have a negative relationship with flowering suggesting that drought in the months of January-February and cloudiness in April could be triggers for flowering in Malaysia.
Graph 11: Scatter plot of climate and flowering in Malaysia
Finally, in graph 12, you can see a relationship between September precipitation and November minimum temperature and flowering in the Philippines. Although it is hard to make any inference for the Philippines data as there were only 4 years of observations, one can see that the negative correlations between the variables are fairly high (rho=-0.95). Flowering in the Philippines was recorded to occur in two main periods: from February to April and from November to December. This analysis suggests that September drought as well as November drop in minimum temperature could be cues for November flowering in the Philippines.
Graph 12: Scatter plot of climate and flowering in the Philippines
In conclusion, the correlations between climate variables and flowering are not very strong for dipterocarp forest in Southeast Asia. However, few common trends have been identified. It seems that drought, clear skies, and drop in minimum temperature have the strongest relationship with flowering in the region suggesting possible cues. On the other hand, different regions seem to respond to slightly different cues (low SOI and precipitation in Borneo, low precipitation and cloud cover in Malaysia, and low precipitation and a drop in minimum temperature in the Philippines). It is hard to make any definitive conclusions about a climatic cue for flowering in a dipterocarp forest based on this analysis. It is possible that dipterocarp trees respond to a resource tracking mechanism that triggers flowering to occur when enough photosynthate has been accumulated. Furhter studies are needed to explore this hypothesis.
