Something a little different to my usual posts and targeted to people interested in Extreme Value Analysis (EVA).
I have been working on an analysis of extreme sea levels as part of my PhD on flooding risk. If you are familiar with EVA in R you will have likely come across the extRemes package, written by Eric Gilleland. While I am not (yet!) an expert in EVA, I have found this package to be quite accessible.
The package’s main function, ‘fevd()‘, fits an extreme value distribution to data. Methods for plotting are available for objects of class ‘fevd‘ to produce diagnostics and result plots, such as a return level plot.
My issue is the restricted control you seem to have over the plot’s layout and general look. You can choose which plots to show, or to add a custom title, but not much else. What if you want to customise the plot to your liking, plot the results in another software/package or display two EVAs on the same figure as shown above?
This post was originally written for the EnvEast Doctoral Training Partnership student blog. EnvEast is a collaboration between three UK universities, including University of East Anglia where I am carrying out my PhD. The blog can be found here.
Fluctuat nec mergitur
“Battu par les flots, mais ne sombre pas. Tossed by the sea, but not sunk.”
From water scarcity to air pollution, without forgetting world-leading greenhouse gas emissions, China is facing a multi-faceted environmental crisis. I was able to witness the extent of the issue first hand when I studied Environmental Management at Tsinghua University in Beijing as part of a joint program with the French university Mines ParisTech. The resounding blow of the August 12th explosions in Tianjin – which killed over a hundred people – has now pulled me back to that time spent in the “Middle Kingdom”. Amidst the apparent lack of available information about the disaster and fears of cyanide pollution, a memory that stands out is my field trip to Tianjin in 2013, which highlights a revealing paradox.