Me, blogging!

thoughts, science, code, jobs, and thoughts

Finish it

Finishing a project is tough. It takes a lot of time. Always more resources than previously anticipated. And of course it is stressful, even briefly before the deadline. And you really start to grasp the meaning of that word: "deadline". There is no afterwards, there is no: Let's do it again! It is the one chance you have been given and you need to take it.

And again, I learned quite a lot abut the last project (cannot tell you more about its content because it's classified, but not in a military or conspiracy sense ;) ). The most important thing again: communication. In many ways it helps to discuss things over and over again. Because there is no time to clarify things afterwards. It doesn't really help to have the aha! moment right after you submit your final report.

Communication is the essential thing for getting things done. It helps to start over, re-think your approach and, of course, to avoid misconception. Because the one simple thing is not necessarily the most obvious thing your counterpart.

The second most important thing: specification. Once you know what exactly to do, it may be easier to follow specific guidelines or requirements. In the best case, those specifications should not change. Specifications help to set mile stones, which in turn help to meet required deadlines.

Last but not least: revision. And revisions take time. So don't start revising your work to late.

The above suggestions do not guarantee a project to be successful but for me they help to structure the project both in terms of time and in terms of resources.

Tags: job 2015/09/18
The Ice Sheet model SICOPOLIS

Recently, I started working with SICOPOLIS, a three-dimensional dynamic/thermodynamic ice sheet model. Download and usage are fairly easy. As long as you want to use the predefined experiments. Although setting up an alternative configuration is not complicated, using you own forcing data is. Nevertheless, the code is well written, the output format is standard netCDF which can be easily processed with—guess what—python and matplotlib.

Here is one example.

Initially, Greenland has no land ice. Under current boundary conditions, i.e. present-day surface temperature and precipitation rates, the ice sheet starts to grow (left figure). After nine thousand years, large parts of southern and central Greenland are ice-covered, mainly because of the large snowfall compared to northern Greenland (right figure).

And I'm amazed. Not because of the results, they seem straightforward. I'm amazed because of the time scales tackled here. Within minutes (literally!), an evolution of a whole ice sheet, which would take many thousands of years, is at my feet. And now I'm becoming aware of the work that lies ahead of me.

In the past 12 months, I was working more or less on the climatology of atmospheric fields like air pressure, air temperature and various radiative heat fluxes. But this was just tens of years. Here, we are talking about time scales 2–3 orders of magnitude larger than this. What an eye-opener. The real question I need to answer is:

How does the seasonal melt cycle affect the large-scale evolution of the Greenland Ice Sheet?

Tags: ice sheet 2013/09/04
About Mario Krapp
I'm a physicist by training and graduated in Earth System Science.

And I like coding. I've been working with complex computer models ever since my undergrad and I enjoy data exploration and data analysis to gain insights into the underlying principles.

Feel free to contact me.