Arctic sea-ice at lowest levels since observations began

RealClimate.org used the R language and data from the National Snow and Ice Data Center to create this chart showing the extent of Arctic sea-ice in each year since satellite observations began in 1978, and the current extent of ice coverage (in red). Even though there are several weeks of annual melting yet to come, the area of ice covering the north pole is at already at a record low:

The record low is the horizontal red line. You can track how far this year's melt will exceed that record by running this R script, which will download the latest data from NSIDC to create an updated chart. (R users may also be interested in the use of readJPEG and rasterImage to create the background for the chart.)

Putting those time series into geographic context, NASA has also released the visualization below, which shows the average minimum extent of sea ice compared with a current satellite view. The yellow line is the average ice pack of all previous years, i.e. the low point of each of the white lines above.

BadAstronomer has more details about how the image above was made.

RealClimate.org: An update on the Arctic sea-ice

Comments

[Halldór Björnsson]

While Rasmus' plot is very good, it can be improved with ggplot2.

Following the

Date[c(FALSE,(diff(Date) < 0))] <-NA

line, remove everything starting with

dev.new(width=10,height=8)
to the end of the file.

Then add the following
--------------------------------------

# Data from Dec 03 1987 to Jan 15 are missing.
# At least Dec 1987 should be removed.

Date[nisdc$Year==1987 & nisdc$Month==12]=NA

library(ggplot2)

# Use 5 day centered average like NSIDC does in the images on www.nsidc.org
nisdc$Ext5d=filter(nisdc$Extent,c(0.2, 0.2, 0.2, 0.2, 0.2),sides=2)

#Stuff Date into data.frame
nisdc$Date=Date

p=qplot(Date,Ext5d,data=nisdc,color=Year,geom=c("line"),group=Year,alpha=I(8/10),main="Sea Ice Extent 1979-2012 by month\n(colored by year with 2012 in black)",ylab="5 day average extent (millions of sq.km)",xlab="Month of year")

pp = p + geom_line(data=subset(nisdc,Year==2012),color="black",size=2,alpha=0.5)

ppp = pp + coord_cartesian(xlim=c(0.01,11.99)) + scale_x_continuous(breaks=seq(1,12,1)) + opts(axis.text.x = theme_text(hjust=2.5))

ggsave("Seaice2102_5d.png",plot=ppp)

------------------------------------------
This yields a plot showing the annual lines in color that starts with blue in the early 80s and changes to red in the last decade