September 17, 2016
The following post is part two of a seven-part series where JSG grad students reveal what they were up to over summer break. Enjoy!
I spent my summer like many final-year PhD students, scrambling to make sure that my results weren’t a fluke. My research involves trying to figure which factors influence the breakup of continents. Continental breakup comes in a variety of flavors: fast or slow, volcanic or non-volcanic, stretched over a long distance or focused over a narrow width. To try to figure out why there are such differences, I conduct numerical modeling experiments. Just prior to the start of the summer, results from one of these experiments led me to a counterintuitive conclusion: if you want to make a lot of melt, cold rocks are better than hot rocks.
To confirm these results, I broke down my code and decided to change how it calculated melt production. So I ended up spending most of my summer going blind in front of a computer screen making incremental changes to the code. Eventually, I had a functional code that could calculate melt by 2 independent methods and could use 4 different published solidi (the pressures and temperatures where solid rock becomes partially molten rock). This essentially gave me 8 independent ways to estimate the amount of melt produced during continental extension. Happily, the results stayed the same: cold rocks clearly make more melt.
To understand why cold rocks are better at making melt, we need to understand how melt is produced during continental breakup; i.e. decompression of the mantle. The style of mantle decompression has a lot to do with the temperature conditions of the lithosphere, particularly the temperature conditions in the lower crust. If the lithosphere is warm, the deeper parts of the crust will behave ductilely. But when the lithosphere is colder, the lower crust will deform brittlely. A brittle lower crust lets faults that form during continental breakup penetrate deeply and focuses deformation. This localized deformation causes the underlying mantle to rapidly upwell and decompress. The mantle decompresses so quickly that the rocks cross the solidus and melt. In contrast, in warm lithosphere faults cannot penetrate through the ductile lower crust. Instead of forming deep faults, we form many shallow faults. These dispersed faults distribute deformation over a wide area and hamper the ability of the mantle to upwell and decompress. The code works, the results make sense, and it fits pretty well with observations from rifted margins. Now I just need to write that paper…
