Step-By-Step Instructions for Creating a Climate-Related Model-Data Comparison Graph

This post is being published in advance of the release of my new book Climate Models Fail so I can link to it in the book.

I had originally included this presentation as a chapter in Climate Models Fail, but the 35 screen captures doubled the file size of the book.  And with ebooks, the price of the book is dictated by the file size. It didn’t make sense to burden the book costs with one chapter, so here it is as a blog post.

NOTE: This post contains about 9MB worth of images.  It may take a while to load, or you may need to refresh the page in order to see all of the images full sized on the webpage.

# # #

Climate Models Fail contains hundreds of illustrations, many of them model-data comparison graphs of surface temperature (land, sea, and the combined land-plus-sea), precipitation and sea ice.

Suppose you see a graph and you find it hard to believe the models stored in the CMIP5 archive performed that poorly. For example, you can’t believe the warming rate simulated by the models that are being used by the IPCC for their 5^th Assessment almost tripled the observed warming rate for the Pacific Ocean over the past 31+ years.   See the following comparison graph in Figure 1. This post provides step-by-step instructions, using 35 annotated screen captures, which walk the reader through the process of creating a model-data comparison graph.  I’ve used EXCEL for the spreadsheet.

Figure 1

The Pacific Ocean is a good example because it straddles the dateline—aka the 180^th meridian and the anti-meridian.  The Pacific stretches from about 120E longitude in the west to about 80W longitude in the east.  It’s also a good example because, as shown in Figure 1, there’s a reasonably large difference between the modeled and observed warming rates.

Both the data and the model outputs are available through the KNMI Climate Explorer, so we’ll use it as our source in the following example.

(You could also use the NOAA NOMADS website for the Reynolds OI.v2 sea surface temperature anomaly data, as I had in Figure 1, but NOMADS only furnishes the data, not the months, so you’d need to pre-format your spreadsheet with a column for time in months.  We’ll avoid that extra step by using the KNMI Climate Explorer.)

PRELIMINARY NOTES

When you’re entering the coordinates of the desired data for the mapping features at the Climate Explorer, they require you to follow a standard format, so it’s best to standardize on one data entry method for everything. That way you know you’re getting the data you desire and not the data for some other region of the globe.

Southern latitudes are input as negative numbers, so, as an example, 60S is entered as -60. Northern latitudes are positive numbers so 65N is input as 65.  Also, the fields used for data entry (Screencap 3) require that the entry on the left contains a lower number than the one on the right.

For longitudes, let’s start with the Atlantic Ocean for an example.  Let’s say you wanted data for the longitudes of 80W to 20E.  For this example, west longitudes are entered as negative numbers so 80W would be -80, and east longitudes (20E) are positive numbers (20).

For entering longitudes, it’s good to visualize a map with north pointing up.  To the left you’d have 80W, so -80 is entered in the left-hand field and, in turn, 20 (for 20E) goes in the right-hand field.   Easy.

With the longitudes of 20E to 120E, the Indian Ocean is just as easy. The longitude to the left on your visualized map is entered as 20 in the left-hand field and the longitude to the right is input as 120 in the right-hand one.  Even easier.

The coordinates I used for the Pacific Ocean are shown in the title block of Figure 1.  They’re 60S-65N, 120E-80W.  Because the Pacific straddles the dateline, the method used to enter the longitudes can get a little confusing. See the map in Figure 2.

Figure 2

The left-hand entry has to be a smaller number than the right-hand entry, so your first thought would be to enter -80 and 120 in that order. But that would get you the data for the Atlantic and Indian Oceans, not the Pacific.

There are two ways to enter the longitudes for the Pacific and they’re shown at the bottom of the map in Figure 2. I use what’s shown as method 1, with 120E input in the left-hand field as 120 and 80W input as 280 in the right-hand one.

One last note:   There are some minor differences (due to the base years used for anomalies) between the graph I’ve presented in Figure 1 and the graph we’ll create in the following step-by-step process. Those minor differences do not change the basic results.  What matters is that the models show that the sea surface temperature anomalies of the Pacific Ocean should have warmed at a rate of that was almost three times that observed warming rate…if the Pacific was warmed by manmade greenhouse gases.

STARTING POINT:  THIS IS A LINK TO THE WEBPAGE SHOWN IN SCREENCAP 1.

There’s more to the post after all of the screen caps.

###

###

###

###

###

###

###

###

###

###

###

###

###

###

###

###

###

###

###

###

###

###

###

###

###

###

###

###

###

###

###

###

###

###

There you have it.  The trends presented by EXCEL are in deg C/year.  For my Figure 1 in this post, I converted the trends to deg C/decade and then revised the graph to my standard format.

Now to confirm something else, delete the months, data and model outputs before 1994. Note the trend line of the Pacific data since January 1994.  It’s flat, indicating the sea surface temperatures there haven’t warmed in 20 years. NOTE: After any change to the data or model outputs, you should always delete the trend lines and have EXCEL recalculate them.

Regards