Some time ago I watched on TV an extraordinary documentary about the last Italian expedition on summer 2007 to the marvelous K2, the famous Himalayan mountain, and amazed by the extraordinary stories around it I started searching for some additional info about it.
For those of you who know nothing about it will suffice to say this:
“Its height is eclipsed only by Mt. Everest, but its level of difficulty is eclipsed by none. One of the Himalayas’ fourteen 8,000-meter peaks, K2 has earned the nickname “The Savage Mountain” due to its violent storms and catastrophic avalanches. Less than 200 climbers have conquered the mountain (compared to 1,700 at Mt. Everest), and more than 50 have died trying.” [Source: Women of K2]
To my surprise, not only I could find plenty of sources about the mythical history of its ascents, but very soon I stumbled upon a whole set of excel sheets with plenty of data to dig into. I was soon thrilled by the idea of visualizing these data in some way and started playing with Excel to find a proper visualization; one that could tell a story.
The chart that I present here is the first result of my exploration, after having tested tens of different visual solutions. To be frank, I’m not totally satisfied yet but this is the best I could create given my experience with Excel and I hope you can provide some useful feedback on how to do it better.
The data represents all the successful ascents of K2 from 1954 (the first ascent) to 2006. Each data item is a single climber with the following associated data:
- Year -> mapped to the x-axis
- Order of arrival -> mapped to the y-axis
- Expedition -> mapped to color (only the first 8 in terms of total number of ascents)
- With oxygen -> marked with an “o”
- Died on descend -> marked with a “x”
- Sex -> females marked with an “F”
Some interesting facts can be extracted from the chart, here are some examples:
- A big gap between the first ascent in 1954 by the Italians and the second in 1977 by the Japaneses … and then some other small gaps. I tried to investigate to find the reason about it but it looks like it is just that despite many attempts, in some years nobody succeeded.
- There are few female climbers in the history of K2 and they are in fact quite famous. Legendary is the “curse on women”: the first 6 female climbers are all dead either on descent or later in other circumstances. In the chart it is easy to spot the three who are dead while descending.
- Oxygen has been sparingly used by a small proportion of climbers. Notably the first ascent without oxygen was on 1978 during the first successful American expedition (From K2 Timeline: “there is some debate about who the first climber to reach the top of K2 without supplemental oxygen was“).
I am sure there are many others that can be spot on the Chart. A timeline of relevant events can be found on k2climb.net
One of the interesting things which are not shown in the map is how different routes have been tried during all these years. I plan to draw another chart where the successful ascents through new routes are shown.
1) An interesting part of the journey that took me from the idea of visualizing the K2 data to a final picture is what I learned by using the Excel chart tools. Before starting I was barely able to create a simple scatter plot with the wizard, at the end I was able to customize every chart I wanted to draw and to create some fairly complex VB scripts. I have contrasting feelings about Excel because there are very good and very bad things about it. For instance, it’s crazy how the default settings for each chart seems to have been designed purposely to create junk. I had to overwrite almost all the default settings to have a neat picture (e.g., changing background from gray to white). At the same time however I am amazed by how flexible Excel is and how it is open to any possibility if a bit of programming is learned.
2) Being a person trained on interactive visualization I had never experienced the process necessary to transform data into a still picture. It’s amazing how a different mindset is necessary to design a visualization in this way (and yet how much basic knowledge on visual perception is needed). Programming a visualization, knowing that a user will be able to interact with it to disambiguate certain information, is easier then when you know that everything must be conveyed in a single image. Related to that is my surprise once again on how limited the visual features are when you want to map data dimensions to visual dimensions. Here in this chart I had to struggle a lot with myself to decide which dimension I wanted to map to which feature. And changing even only one of these mappings can change the whole story.
3) Even if I agree with many on saying that Excel is a mess and that obtaining the desired results is a pain, I had a very bad experience trying with other software. I’m totally shocked about it! Before I was convinced that Excel was the best way to go, I downloaded tens of little and big applications full of features and all promising to be a piece of cake when creating charts. To my experience this is plain false. The best alternatives are represented by complex software like SPSS or Illustrator which in many ways are better than Excel (e.g., Illustrator is great if the final result must be manipulated by hand and it’s meant to be printed) but still very hard to use and to learn (and very expensive too!).
4) I think there is a full potential for interesting investigations out there about sport and expeditions datasets. The main data set I have used here was extracted from a web site full of data on many kinds of adventures. See Adventure Stats for more details. I am sure that visualization can help telling many interesting stories about exploration in compact and well designed charts.