Cloudy with a chance of modelling!

ac307
Thursday 25 July 2019

 

Upon starting my project I had to decide what my first steps would be. I had many papers to read and masses of code to understand before I could begin to add my own touches which was a daunting task. I felt that in five weeks such a task would be nigh on impossible! So I started, head first, hitting the ground running and before I knew it I was crashing and, most definitely, burning. I had hit my first obstacle. The code I was reading was unlike anything I had seen before and in an entirely new language that was wholly unfamiliar to me.  I felt defeated from the get-go, how am I going to get all this, a project that has been in the planning for almost 3 months, done in 5 weeks? I persevered and after various frantic meetings with my supervisor, and a couple sleepless nights, we developed a plan to get through the initial stages of my project.

I suppose now would be a good time to explain what exactly I was getting up to in those five weeks. My supervisor, Professor David Dritschel, is currently working on an experimental computer model which combines two very well tested methodologies in computational fluid dynamics, namely the Eulerian model type (which assigns fluid property values to a grid, evolving the fluid relative to the grid and taking contours) and the Lagrangian model (which evolves fluid properties relative to moving parcels). The model intends to be a ‘proof-of-concept’ that one can create a hybrid model which would speed up computation times significantly. The model simulates the ascension of a plume of warm liquid-laden air that, due to its high water content and temperature relative to its surroundings, is highly buoyant. This causes the plume to rise and as this plume rises it will begin to spread out in a thermodynamic process known as diffusion. The plume will also begin to lose heat, cooling down the packets of moist air. The culminating result of all this is the eventual occurrence of condensation once the plume passes what is known as the dew point, a theoretical point where the plume would begin to lose water content and thus lose buoyancy. My job was to assess the effect of an atmospheric phenomenon known as environmental shear, this is where there is a sudden arrangement of wind (or currents) which line up in such a way that as height increases the air movement progressively gets stronger in a single direction and in a linear fashion or in other words if the height doubles as does the speed of the wind. After this so-called linear shear was added I was to then test the effects of a rotational shear, changing parameters such as strength and rate of rotation to gauge the overall end result on the progression of cloud formation.  Safe to say this, along with the aforementioned initial stages, made up the first two weeks of project.

The third week was a welcome change from the usual coding and reading as I was given the chance to attend and observe an international seminar on atmospheric fluid dynamics! This was by far one of the main highlights of my project. It was a fantastic experience to be able to see the cutting-edge research being done in this field and also to rub shoulders, meaning annoy with my undergraduate level questions, with eminent fluid dynamicists. The talks were an incredibly valuable experience as I got the chance to see how these professors and post-docs wielded their public  speaking skills, a skill I have struggled with since very young, and most importantly how they handled difficult questions from an eager and well-informed audience.

 

Colour mapping indicates liquid humidity level
A plume rising without shearing forces

After implementing the shears to the code I noticed that there was no way to fully visualize the results of the code in full 3D which was slightly frustrating as I found it hard to explain exactly what the effects of the shears were to the cloud formation. So I approached my supervisor and asked if there was such a tool already built-in to the existing and code and in-fact there was not! I knew this was a fantastic opportunity to add something to the package wherein its usefulness would extend beyond the code it was designed for. So I got to work, spending the final two weeks of the project testing different approaches to this problem and after many failed attempts I finally found a method that clearly demonstrates the progression of the plume and shows, in reasonable detail, the formation of a cloud. Both I and my supervisor were pleasantly surprised at how well the results came out in the shiny new 3D visualization. I was left feeling very accomplished and very much relieved that all the effort I put in had paid off; I am sure my fellow Laidlaw cohort would agree with the following sentiment – it is a great feeling!

A plume rising with shearing forces

I would now like to take the opportunity to reflect on the project and its overall impact. I believe that I have learned a huge amount about academic research and leadership while being given experiences I won’t soon forget. I got the chance to see how this kind of research is done and it has put me in contact with people I would not have otherwise had the chance to meet and for that I am immensely grateful. I have to say that the experience was also very humbling as it taught me that research and leadership (as a developing skill) isn’t as easy as I previously had thought, it takes serious commitment and sacrifice. I definitely came away from this experience with new found respect for those in research positions.

Finally, I would like to say that the main take-away thought I had from these first five weeks is that no matter what you are doing, no matter how hard or complicated, is to always persevere and to love what you do.

Share this story


Leave a reply

You must be logged in to post a comment.