jssailem
SBO Weather and Forecasting Forum Jim & John
- Oct 22, 2014
- 22,825
Those scientists who are riding a boat up in the Arctic are going to have ice up-to their keisters.
high water?
- Thread starter jon hansen
- Start date
@JamesG161 @jssailem
I hope I’m not coming across as a doomsayer - just curious. Although I can see the more aggressive trend line for this year I also note that it’s still below the average amounts. Do you see this year bringing things back to the average or surpassing it?
I’m OK with a cold winter, reminds me of my youth when we could safely drive cars on the ice and didn’t have to wait for the ferry.
I hope I’m not coming across as a doomsayer - just curious. Although I can see the more aggressive trend line for this year I also note that it’s still below the average amounts. Do you see this year bringing things back to the average or surpassing it?
I’m OK with a cold winter, reminds me of my youth when we could safely drive cars on the ice and didn’t have to wait for the ferry.
Crazy Wolfe Islanders!
@JamesG161 For us Great Lake sailors it would be interesting to see a comparison of lake levels and precipitation. Do you have an easily accessible source for that?
Doomsayer, not a bit. The scientific world is nearly unanimous: Sea level is rising at about a foot a century, and has accelerated. There is hard evidence that human activity is adding to the acceleration. This is known.@JamesG161 @jssailem
I hope I’m not coming across as a doomsayer - just curious. Although I can see the more aggressive trend line for this year I also note that it’s still below the average amounts. Do you see this year bringing things back to the average or surpassing it?
I’m OK with a cold winter, reminds me of my youth when we could safely drive cars on the ice and didn’t have to wait for the ferry.
The most worrisome unknown is how fast it will continue. I worry too about your grandson and mine, that aren't here yet.
Those of us in the developed world are managing to keep our low lying populated areas above water. It's the areas you show in your post, much in undeveloped countries, that are at the highest risk.
It's sad to say but the best thing you can do for your grandson regarding sea level rise, is done; he's born into the first world.
Weather is another matter: I'll take a shot, I'm predicting a pretty,...regular winter.
I've ordered my usual 3 tons of wood pellets but I haven't started the pellet stove yet as the heat pump is supplying all our heat needed at as much as 300% efficiency (compared to an oil furnace @ 90% efficiency). I didn't invest in these systems because they're green, I did it because I save money.
And the heat pump was nice in reverse (AC) this past July, that was in fact the hottest month ever recorded on earth.
Not crazy all of our Mom’s had us tested!
Check out the forecasts at the bottom of this page.
Kingston, ON - 7 Day Forecast - Environment Canada
Current conditions and forecasts including 7 day outlook, daily high/low temperature, warnings, chance of precipitation, pressure, humidity/wind chill (when applicable) historical data, normals, record values and sunrise/sunset times
This is Great Lake ice coverage trend. Note the whipsaw - lowest to highest in recent years.
Interesting graph. The median coverage hasn't changed but the variability has; with greater variability comes less predictability. The more energy in the system(atmosphere) the less stable the atmosphere becomes.Not crazy all of our Mom’s had us tested!
Check out the forecasts at the bottom of this page.
Kingston, ON - 7 Day Forecast - Environment Canada
Current conditions and forecasts including 7 day outlook, daily high/low temperature, warnings, chance of precipitation, pressure, humidity/wind chill (when applicable) historical data, normals, record values and sunrise/sunset timesweather.gc.ca
This is Great Lake ice coverage trend. Note the whipsaw - lowest to highest in recent years.
View attachment 171713
It is also instructive about making predictions based on too few data points. The flat line for the median ice coverage would suggest that climate change isn't occurring or is having no effect on ice coverage. However, the increase in variability suggests that something has changed in the past 30 years.
My opinion is that human impact on natural cycles is real and significant.
Some argue that 30 years is just a tiny hint of a blink in geological terms so do we really have enough data to definitively say what is going on.
@TomY made a interesting point about the reason for change in how we go about things is often driven by financial concerns. There is lots of data that suggests money managers are going long-short on company's that are green vs not.
The only real power I have is over my own little domain. Last year I found I used our sailboat a lot more often than our power boat, wind is free, gas is $1.70/litre at our marina.
Two points here.
1) Sea Level Rise
2) Excess Fresh Water in Great Lakes.
I have done a lot of research on Point 1.
But first you need reference point as a Standard to compare.
Climatology has picked a range of dates to do the Average of the period 1981-2010 [Satelite observations era]
Point 2
A Dam Controlled or Flood dump [Mississippi River] level is where Man decided his reference point.
Obviously More Input and Less output, Levels Rise.
Climate decides the Rates of Input of water.
Point 1
Here is where we are still leaning. Several reference points are changing. We found Land is sinking in some parts of the World. Plate tectonics.
So most agree we have to watch the Poles for ICE changes and that has just really started, using Satellites.
My post on ICE is a new system just started for best estimation and monitoring
____
Last point...
Is our SUN...
It is now changing to a point we have never seen before. Solar Storms are way down. I will leave it there for now, but you might research around about the current State of our Sun.
Jim...
PS: Careful on picking reference points.
1) Sea Level Rise
2) Excess Fresh Water in Great Lakes.
I have done a lot of research on Point 1.
But first you need reference point as a Standard to compare.
Climatology has picked a range of dates to do the Average of the period 1981-2010 [Satelite observations era]
Point 2
A Dam Controlled or Flood dump [Mississippi River] level is where Man decided his reference point.
Obviously More Input and Less output, Levels Rise.
Climate decides the Rates of Input of water.
Point 1
Here is where we are still leaning. Several reference points are changing. We found Land is sinking in some parts of the World. Plate tectonics.
So most agree we have to watch the Poles for ICE changes and that has just really started, using Satellites.
My post on ICE is a new system just started for best estimation and monitoring
____
Last point...
Is our SUN...
It is now changing to a point we have never seen before. Solar Storms are way down. I will leave it there for now, but you might research around about the current State of our Sun.
Jim...
PS: Careful on picking reference points.
How about less Energy input to the System?
I would avoid Stability since that appears to be a localized result.
Remember we are a WATER planet. Our Oceans Stabilize our Atmosphere in general. Land does also, but that is what we see changing by man.
Jim...
No more than the Mississippi River and its tributaries.
Here is where we have local observations. Before man took to the airs, local observations were very limited, Globally.
When man became airborne, the airports were the single spot for Weather data.
Now we have "weather stations" in our back yards.
Here is the ironic part that data is "manipulated".
When did the airports read their DAILY data? Midnight to Midnight? Day break to day break?
So NOAA tries to adjust the readings and compare against history to find the HIGHEST and LOWEST temperatures.
Who cares really other than the News Cycle?
We are reading AIR temperatures in urban and rural areas.
Did you change your day based on AIR temps more than 1°F change? or 10°F?
Jim...
PS: The non-adjusted NOAA data show we had the coolest weather in North America for the time of year.
If we have Doom coming, it won't be stopped by man.
Averages are really dampened looks at things, but they do compare. Remember being graded on the Curve in school?
The teacher used Averages to make a C and compare to others in the class.
Did that really mean you were Average intelligence? Or that the Teacher's test sucked?
____
Trend lines are statistics. Our input data is Zillions time more than the past. Careful on trend lines too.
Weather patterns tend to repeat every 60 years. Why? Oceans.
Jim...
Statistic and trends....
Note the Graph on my post#19
At the bottom they say the GREY area represents ±2 Standard Deviations[SD]. That assumes a bell shape curve like teacher's grading.
±2 SD is about 97% of the data collected.
Note the spread during the Summer versus the Winter.
So are we going to be in the ±2 SD this year or perhaps higher than Average?
Jim...
Note the Graph on my post#19
At the bottom they say the GREY area represents ±2 Standard Deviations[SD]. That assumes a bell shape curve like teacher's grading.
±2 SD is about 97% of the data collected.
Note the spread during the Summer versus the Winter.
So are we going to be in the ±2 SD this year or perhaps higher than Average?
Jim...
Assuming a normal distribution, ± 1 standard deviation represents about 68% of the distribution, ± 2 SD represents about 95% of the distribution. (There is about 2.5% above 2 SD and about 2.5% below SD.) So when a prediction made based on 2 SD it really isn't much of a prediction, however, if an event occurs that is greater than 2 SD above the mean, then the chance of that occurring is about 2.5%, which is pretty rare.
The chart Jim references shows that the ice coverage is way less than 2 SD below the mean. If we were to compute a test of statistical differences, there is little doubt the reduction in ice coverage is statistically significant and highly unlikely that it is due to random fluctuations. In other words, there would be less than a 2.5% chance of the low ice coverage being "normal" and random.
Results of applying principles of statistical analysis to an extremely small time based sample of a slowly changing historical parameter are almost certain to be contaminated by randomly occurring short term events. An analogy might be estimating the time based average speed of cars on a highway by measuring the speed of one car over 1 mile.
Well, yes, however, the graph Jim shared showed the time sample over 29 years and showed the current year being considerably outside the 2 SD band for the 29 year period. In the social sciences an N of 30 is generally considered adequate for a lot of research; so, I think it is safe to say the ice coverage thus far this year is significantly lower than than the base (control) years. Now that says nothing about cause and effect and it may or may not be a trend. It does say the probability of the low ice coverage being random is quite low, somewhere around 1%. It could be random, but that's nothing I want to bet on.
A better example of you point is the graph in post #25. It is clear from visual inspection there is a lot of variability in Lake Ontario ice coverage, however, over the course of 45 years the median has remained unchanged. Analysis of that trend requires a different set of statistics, i.e., is the increased variance statistically significant? Unfortunately, my last course in multivariate statistics was in 1989 and in the intervening years doing the calculations has not been part of my career, so I'm a little limited on delving deeper into the nuances of an analysis.
Regardless of the long or short term trends, the data suggests increased variability and with that increased uncertainty about future weather patterns. This was the nature of a discussion I had with the owner of my marina this afternoon as he considered his (very expensive) options for hardening the marina against future flooding.
Don't you know 86% of all statistics are made up on the spot?
Thanks for the correction, Dave.
≈97% is ±3 SD.
I was so accustom to Quality Control of products to be better than ± 3 SD.
In class room curves, the High side would be the "A" and the low side "F".
_____
Not if you have a model that can extrapolate and has been tested against real data.
______
Ok you trend watchers...
Time to take your bets on this winter's Arctic Ice for Winter of 2019.
I would suggest we look again on Jan 15, 2020, to declare the winner.
This graph lags by a day. So this it for Nov 5, 2019
If you project the Ice trend, which most say will lower the Oceans if the poles gain more ice,
Will it be Above the 1981-2010 median or Lower ?
My bet is Above
Jim...
Something at the 97.5 percentile is 3 standard deviations above the mean. Something 3 SD below the mean is at the 2.5 percentile. 95% of the scores/measurements are between the 3 and -3 SD from the mean. 5% are outside the ±3 SD band, 2.5% below and 2.5% above.
An easily confused concept. As a school psychologist I live by the normal distribution, it is my mantra.
Notice on that new graph they introduced a different Statistics...
Now Interquartile and Interdecile.
They really trying to parse up the distribution of data points.
I am not so sure their data is that accurate.
In true sampling and data taking, they should also show the number of samples taken and that Standard Deviation of sampling.
But...
This is a relatively new presentation so I will go with the flow.
_______
What no bets on the trends in my post#37?
Chicken?
Jim...
Now Interquartile and Interdecile.
They really trying to parse up the distribution of data points.
I am not so sure their data is that accurate.
In true sampling and data taking, they should also show the number of samples taken and that Standard Deviation of sampling.
But...
This is a relatively new presentation so I will go with the flow.
_______
What no bets on the trends in my post#37?
Chicken?
Jim...
WTF! Well that sent me to Google.Notice on that new graph they introduced a different Statistics...
Now Interquartile and Interdecile.
They really trying to parse up the distribution of data points.
I am not so sure their data is that accurate.
In true sampling and data taking, they should also show the number of samples taken and that Standard Deviation of sampling.
But...
This is a relatively new presentation so I will go with the flow.
_______
What no bets on the trends in my post#37?
Chicken?
Jim...
So the interquartile range is the range between the 25th percentile and the 75th percentile incorporating about half the distribution. In psychology we typically call this the average range as half of the data falls within this range.
The Interdecile range is bounded by the 10th and 90th percentile or about 80% of the data points.
Loosely interpreted, any value within the interquartile range would be considered average or typical in the common vernacular.
Likewise, any value outside of the interdecile range would be considered an outlier or atypical, as there would be a 20% chance of the data falling outside that range (10% on either side). In the weather and ice data we would consider that unusual and perhaps noteworthy.