Lowest and earliest Arctic sea ice maximum

PIOMAs northern sea ice volume

The northern sea ice has now reached its seasonal maximum extent. This year’s is the lowest and earliest maximum of the satellite era (since 1979).

. . . → Read More: Lowest and earliest Arctic sea ice maximum

Trend significance

Simulated Spencers Creek peak snow depth slope probabilities with nil trend

Updated: happy now; the words “hypothesis” and “null” do not appear (except to discard them).

There’d be no more abused tool in all of science than the linear regression “p-value” for trend significance. Don’t take my word for it; the problem is so severe that some technical journals have actually considered banning publication of p-values.

. . . → Read More: Trend significance

Spencers Creek snow depth prediction model - mark IV

Prediction_v_performance_MkIV

Updated: added residuals plot

I’m really impressed by this. I now have a snow depth prediction model that “explains” 50% of the variance in the Spencers Creek season peak snow depth record (midway between Perisher Valley and Thredbo, from Snowy Hydro). When you consider the vagaries of weather, snowfall, compaction, melt and weekly measurement, that really is a surprising achievement. I sure didn’t expect to get anywhere near it when I started out nearly a decade ago.

. . . → Read More: Spencers Creek snow depth prediction model – mark IV

Volcanoes and snow

Precursor eruption of Pinatubo, June 1991 (USGS)

A precursor eruption of Pinatubo, June 1991 (USGS)

Large volcanic eruptions can affect global climate — ask the folk who starved to death in Europe in 1816, after the massive eruption of Tambora in faraway Indonesia the year before. Explosive eruptions often inject vast quantities of ultrafine material high into the stratosphere, where it spreads around the planet blocking incoming sunlight.

. . . → Read More: Volcanoes and snow

Sunspots and snow

Sunspots_since_1750

Bear with me here. I come from a place with a long history of nutty sunspot-based weather prediction, so I’m well aware of the pitfalls. But our sun really is a very slightly variable star, and aspects of our climate really do very slightly vary with its output — particularly snow cover.

. . . → Read More: Sunspots and snow

The frequency of 'records'

Simulated future record temperature years

If you begin monitoring some variable thing over time, initially you’re going to see lots of new extremes — new record highs, for example. As your dataset grows, the frequency of new ‘records’ should decline sharply. If the overall system behavior is static, new records rapidly become very rare indeed, governed by the statistics of the variation. A feature of the 165-year long global temperature series in recent years is that new records highs have not been rare at all.

. . . → Read More: The frequency of ‘records’

2014 the warmest ... again

Bloomberg monthly NOAA NCDC animation

Update: Added note on Berkeley Earth.

Both the NASA GISTEMP and NOAA NCDC global temperature series have updated for December, confirming that the year just completed was once again the warmest on record. That’s since 1880 when those series start, but really since at least 1850 when the other instrumental series begin

. . . → Read More: 2014 the warmest … again

Landsat snow

Snow extent 1977 - 2014, New South Wales snowfields, Australia

Another way to look at snow cover is by satellite remote sensing. The longest and best-known series by far is that from the US Landsat satellites, now up to Landsat 8, launched in early 2013.

. . . → Read More: Landsat snow

More thoughts on global temperature

Simplistic temperature extrapolations

Having taken the trouble to plot all eight popular global temperature series together on one graph at monthly resolution — something the other seven billion of you don’t seem to have bothered with — it may be fair to spare us the indulgence of a few simple observations

. . . → Read More: More thoughts on global temperature

Fitting the Pearson type 3

PIII_fit

You’ve got some data and think a Pearson type III distribution might fit it nicely, but how do you go about choosing the parameters? The obvious way — using the mean, standard deviation and skewness of the sample — is much frowned upon. That’s because it can give a biased fit, although in the real world it often performs well, as we’ll see. . . . → Read More: Fitting the Pearson type 3