You'd think NASA would know well enough to start the y-axis at 0. Lousy graphs of this sort create the appearance of deceit, even in it absence, and wind up giving ammunition to skeptics.
Why does the y-axis of a graph need to begin at 0? 0 is a number like any other number. If the result set you're attempting to display doesn't have any values at 0 then why would you display 0? The graph in question also doesn't show any data points at y=-257,687 ... does that concern you?
In general, in a time-series, use a baseline that shows the data not the zero point. If the zero point reasonably occurs in plotting the data, fine. But don't spend a lot of empty vertical space trying to reach down to the zero point at the cost of hiding what is going on in the data line itself. (The book, How to Lie With Statistics, is wrong on this point.)
For examples, all over the place, of absent zero points in time-series, take a look at any major scientific research publication. The scientists want to show their data, not zero.
The urge to contextualize the data is a good one, but context does not come from empty vertical space reaching down to zero, a number which does not even occur in a good many data sets. Instead, for context, show more data horizontally! .
> The book, How to Lie With Statistics, is wrong on this point.
Ugh, the more I read of Tufte the less I like him. He loves to make blanket statements (like calling a point "wrong") in an authoritative way, and he's amassed enough adoring followers that people repeat his words like gospel instead of calling him on it.
It's valid to say that a non-zero-based scale shows the data better. But it's also completely valid to note that non-zero-based scales can be used in alarmist ways to make data extremely misleading.
Because when zero is not marked it is impossible to judge the relative magnitude of all graph fluctuations without looking at the y axis scale. On this graph a lot of people (incl. me) would be tricked into thinking that in the last 100 years the amount of CO2 increased by an order of magnitude. Now, I know that's impossible, and I checked the scale at the left to find the usual suspect. But many normal people would just believe what they see. It's typical deceit method that is widely used in corporate presentations, journalism, etc. The only excuse for not marking zero on most charts is if you want to show minor fluctuations more precisely, which is not the purpose of this graph.
The magnitude of the fluctuations should really be judged against the error bars, not against zero. That's what tells you whether the change is significant.
But that doesn't tell you whether it's important. Statistical significance is a measure of trustworthiness, given assumptions about the generating process. It tells you nothing about whether the data is in any way meaningful, after assuming relibility.
What does tell you about that type significance of (rational) data is the delta, which requires the 0 in order for you to see the scale. Now, in many interesting cases this could be represented as error bars under H0, but that's usually not done in publications, because they only have room for one plot type.
The range of the plot is not arbitrary. To suggest otherwise is deceptive.
When you're looking at a line chart that starts at zero, it's very difficult to really understand what's going on unless there's a really huge amount of variability in the data. For example, charts of stock prices virtually never start at zero. Does that mean that, say, Google Finance is trying to deceive you into thinking that Apple's price went up ten times today? No, it means that they're assuming that you're not stupid and that you know how to read a chart. NASA very clearly labels the chart, and anyone who isn't astoundingly stupid or willfully trying to misunderstand the data can see very clearly that this chart starts at 160.
0 is not like any other number. It is of very special import, along with 1. In particular, when dealing with 2D graphics, people look at area. Area is defined by multiplication, and 0 and 1 are the two numbers y for multiplication such that f(x,y)=y and f(x,y)=x for all x.
0 is the only number that is not arbitrary in a scalable measurement such as a graph. That's what makes it important.