In a previous post, I tried measuring the global snowfall trend over the last 41 years using a pixel color technique because I couldn’t find the original data behind NASA’s public images. I have now found the higher resolution data needed to find the most accurate global snowfall trend.
The data is from here1. The results are very similar to the previous result, thus the pixel color technique was excellent.
Here is the real global snowfall trend:
Linear Regression Trend: From 2.855 To 2.949 is +3.292%Global snowfall has increased by over 3 percent in the last four decades.
By hemisphere:
Linear Regression Trend: From 2.771 To 2.538 is -8.415%
Linear Regression Trend: From 2.939 To 3.358 is +14.232%That’s all. Enjoy 🙂 -Zoe
Code
# Zoe Phin, 2021/11/25
# File: snow.sh
# Run: . snow.sh; require; download; snow; plot
# Data: https://goldsmr4.gesdisc.eosdis.nasa.gov/opendap/MERRA2_MONTHLY/M2TMNXFLX.5.12.4/contents.html
require() { sudo apt-get install -y gmt gnuplot; }
download() { user=username; pass=password
let n=1; for y in {1980..2020}; do for m in {01..12}; do
[ $y -eq 1992 ] && n=2; [ $y -eq 2001 ] && n=3; [ $y -eq 2011 ] && n=4;
url="https://goldsmr4.gesdisc.eosdis.nasa.gov/opendap/MERRA2_MONTHLY/M2TMNXFLX.5.12.4/$y/MERRA2_${n}00.tavgM_2d_flx_Nx.$y$m.nc4.nc"
wget -O M$y$m.nc --user=$user --password=$pass -c $url
done;done
}
one() {
ncks -HC --trd -v $1 M$2$3.nc | awk -F '[= ]' -vy=$2 -vm=$3 '{
n=2; if ($4 == "-90" || $4 == "90") { n=4; $4=89.875 }
a=6378.137; e=1-6356.752^2/a^2; r=atan2(0,-1)/180;
A=1.25*(a/n*r)^2*(1-e)*cos(r*$4)/(1-e*sin(r*$4)^2)^2
SA+=A; S+=$8*A
} END { printf "%.3f %.3f\n", y+m/12-1/24, 1e6*S/SA }'
}
snow() { for y in {1980..2020}; do for m in {01..12}; do one PRECSNO $y $m; done; done | tee snow.csv; }
plot() {
cat snow.csv | gmt gmtregress | awk 'NR>1{print $1" "$2" "$3}' > snow.dat
sed -n '1p;$p' snow.dat | awk 'NR==1{S=$3} NR==2{E=$3} END { printf "\
Trend: From %.3f To %.3f is +%.3f%\n", S, E, (E/S-1)*100 }'
echo "set term png size 740,470; unset key; set title 'Global Snowfall (mg/m^2/s)'
set grid ytics xtics; set mxtics 5; set mytics 5; set ytics format '%.1f'
set xrange [1979.5:2021.5]
plot 'snow.dat' u 1:2 w l lw 1 lc rgb '#0000EE',\
'snow.dat' u 1:3 w l lw 2 lc rgb '#000088" | gnuplot > snow.png
}
Note 1: This data requires user registration here. Replace username and password in the code.
Zoe's Insights 
Thanks Zoe, nice one, as usual.
LikeLiked by 1 person
Absolutely breathtaking! A reproducible result requiring only some ability and interest in facts. Oddly enough, only 1/9 of the human population lives in the Southern Hemisphere. This order-of-magnitude difference make one wonder if “we” ought maybe to give up on bullying women into the involuntary labor of reproduction, rather than try to return to bucolic medieval ways with 9 billion people (double the 1971 population) burdening the planet.
LikeLike
I see overpopulation as an impetus to get off this planet. I think those that want to keep the population low are dooming us to a cataclysm that will destroy all human kind forever. We need to get off and multiply everywhere.
LikeLike
Women love to bear and look after children. Most men enjoy fatherhood. However, birds and humanity alike, prefer to have a ‘nest’ to occupy before they start a family.
It’s observed that the birth rate declines as income and wealth increases.
Where income and wealth goes to an ever diminishing fraction of the populace the struggling bulk of humanity finds it difficult to set up the ‘nest’.
In developed societies like Australia, Japan and the large cities in China, the birth rate is now well below maintenance levels. Asset prices have been inflated beyond the reach of the common man. Strangely, many homes remain unoccupied. This is the situation in many countries.
Its quite common for those who have the wherewithal, to observe that the ‘strugglers’ are far too numerous (overpopulation). The wealthy live in leafy suburbs and they hate the idea of high rise and social housing. Town planning was developed to look after their interests. Local Authorities that are in charge of town planning act on the basis of complaints. It is the ‘haves’ who complain, not the ‘have nots’ who are convinced they are unworthy.
Humans are self obsessed creatures. Mothers and fathers, of necessity, get training in being ‘other focused’. Having children is a civilizing process.
The planet is producing well below capacity. The current boom in manufacturing in China, the logistical problems and the inflation tells you that when the government gives funds to middle and low income earners, who have pent up demand, that money is spent immediately. Ideally, its funded by taxation. Borrowing creates problems for later.
The problem is not ‘unsustainability’ due to an imbalance between resources and population, but the unwillingness to share income, wealth and space. Those who are most obsessed with money are called ‘financiers’. Here the income stream is strongest when ‘rorting’ the system. Tax dodges are about milking the flow that would otherwise go to those who need the funds most. A potentially very large flow, turns into a trickle. There are lots of snouts in the tax avoidance trough.
Zoe, what’s your perspective, given you experience as a mother, and on Wall Street? What’s your perspective on borrowing? What’s the future of the US? The $US as a medium of international exchange? The future of the imperialism?
LikeLike
All I can say is that the Anglosphere is in decline.
I’ve invested heavily in commodities, crypto, software, robotics/automation, telehealth, and bargain retail. Stuff like that.
Sorry, I still haven’t found time to read your book.
LikeLike
Another great analysis Zoe. The Hemisphere difference is a big surprise, I suppose it goes with Antarctica cooling.
LikeLiked by 1 person
Hey Zoe,
What’re your confidence intervals for those trends?
LikeLiked by 1 person
Good question. Mine is 100%, ha ha, but I didn’t read about the underlying confidence of each data point. If climate scientists wanted less snow, they had ample chance to do it here. This snow trend also matches general precipitation trend.
LikeLike
Oh I wasn’t clear! I meant this type of confidence interval:
https://en.wikipedia.org/wiki/Confidence_interval
I always like to know how uncertain results are when weighing evidence.
LikeLike
Yes, I understood that. Same answer.
LikeLike
So your confidence interval is +/-100% per… year? decade? Full record?
LikeLike
I didn’t have time to read the manual. Wasn’t that obvious? 95% is standard in NASA products, but some can go as high as 99.5%. I don’t know about this particular product, nor do I really care. I don’t care because I don’t see anything better. Period.
LikeLike
Ah I think I see the confusion.
I meant a trend confidence interval as in what you calculate from the data. What are your standard errors on your trends?
LikeLike
None. I simply plot the best linear fit (smallest variance) on data that doesn’t come with any intervals. The data is opaque on this question.
LikeLike
I do the same thing. ….
LikeLiked by 1 person