Rgbrock1:
A pressure drop of that magnitude would have to be bounded by a funnel cloud. Tornadoes are atypical in the tropics, and one open all the way through the tropopause rarer still.
There are ample, more common scenarios resulting in a steep climb, the result of the automation's algorithms and the unfortunate coincidence of vertical and or horizontal wind shear.
Soon the FDR data will reveal the various parameters, attained values and consequent flight path. I am not expecting anything too earth-shattering to be revealed. What the industry and regulators conclude from the data is another thing entirely.