|
Post by mondeoman on Sept 12, 2017 22:01:02 GMT
But it's the warmest snow Eva.
|
|
|
Post by Ratty on Sept 12, 2017 22:29:39 GMT
But it's the warmest snow Eva. I had a fried ice cream dessert once .....
|
|
|
Post by nonentropic on Sept 13, 2017 3:04:39 GMT
yes snow in the south of NZ on the mountains.
Looks quite 70's to me in the style of winter in play.
|
|
|
Post by nautonnier on Sept 13, 2017 11:43:55 GMT
But it's the warmest snow Eva. I had a fried ice cream dessert once ..... Freeze ice cream solid, take solid ice cream and cover in a batter, deep fry until batter is golden. Eat and put teeth into thermal shock. Scots have been doing this for years with 'Mars Bars' - same approach.
|
|
|
Post by Ratty on Sept 13, 2017 13:04:48 GMT
I had a fried ice cream dessert once ..... Freeze ice cream solid, take solid ice cream and cover in a batter, deep fry until batter is golden. Eat and put teeth into thermal shock. Scots have been doing this for years with 'Mars Bars' - same approach. I thought it was Iceland's national dish ........
|
|
|
Post by sigurdur on Sept 13, 2017 14:46:03 GMT
We cook it in cod liver oil to add flavor!
|
|
|
Post by missouriboy on Sept 13, 2017 14:59:11 GMT
We cook it in cod liver oil to add flavor! I thought it sounded fishy. Must be a longitudinal climate thingy.
|
|
|
Post by slh1234 on Sept 13, 2017 16:31:13 GMT
In conversations with a data scientist today, I heard what might have become my new favorite quote. That is: All models are wrong. Some are useful. (He was talking about financial models, not climate models, but it applies either way. )
|
|
|
Post by nautonnier on Sept 13, 2017 16:39:50 GMT
In conversations with a data scientist today, I heard what might have become my new favorite quote. That is: All models are wrong. Some are useful. (He was talking about financial models, not climate models, but it applies either way. ) Well they can be useful as they allow you to do sensitivity analyses to certain parameters which tells you a lot about the model and/or your understanding of the parameters
|
|
|
Post by slh1234 on Sept 13, 2017 16:58:22 GMT
In conversations with a data scientist today, I heard what might have become my new favorite quote. That is: All models are wrong. Some are useful. (He was talking about financial models, not climate models, but it applies either way. ) Well they can be useful as they allow you to do sensitivity analyses to certain parameters which tells you a lot about the model and/or your understanding of the parameters In his case, the meaning was a bit different, but the cute expression still applied. He was presenting his use of AI in financial forecasts. Of course, it is never exact, but in analyzing the range of errors, where you want to see errors as a tall, narrow distribution very close to the line that represents the actual prediction/projection, it showed that the AI projections his team had produced over the last few years were more consistent, and more accurate than the human-derived projections. He was not promoting AI in place of humans, but rather an AI + human mode of producing projections (Because of human bias on one side and the events that humans will know that cannot be represented in the data on the other side.) So in his case, the model is not exact, but it is more accurate than the human-alone predictions, and so the model is useful. Another aspect is given in the example of probabilities that you will survive the wreck of the titanic based on your cabin. There are models and data that are distributed in the open-source R libraries where you can project that. But why? It's useless as a business problem. So two angles encapsulated in his point. I just really liked clever the expression of it.
|
|
|
Post by missouriboy on Sept 13, 2017 17:44:26 GMT
Post-Nino Analogs For 2017 Weather in Columbia, MO
I have been making much of the 2017 cool weather pattern we have experienced here in Columbia MO since the end of July. This is, of course coincident with what appear to be similar weather patterns in other areas of the planet. So I decided to see if I could find similar analog years that had the same general pattern ... and it occurred to me that a comparison to past post-NINO years might be informative. This is simply visual ... no complicated statistical algorithms. The four charts below show running weekly means of maximum and minimum daily temperatures and their deviations from the running weekly means for the 1981-2010 normal period. I focused on 1999, 2011 and 2017, which are placed at similar distances following the 1997-98, 2009-10 and 2015-16 ENSO events, forgetting for the moment the question of whether the 2015-16 event was equivalent to the other two, given that it post-ceded solar cycle 24 maximum versus the other two that preceded the maximums of solar cycles 23 and 24. All three years follow a very similar trajectory with similar monthly-seasonal patterns. The largest difference is that the relative cold beginning these years extended into mid-February in 2011, whereas the 1999 and 2017 cold period ended in mid-January, remembering that 2011 had a relatively strong negative NAO. This aside, winter-spring temperatures and their associated deviations were very warm. Deviation charts show that temperatures were generally at or above the 90th percentile of weekly norms during this period and steadily declined to the boundary of the 10th percentile into September (where we are now). During October, both 1999 and 2011 moved back toward the 90th percentile upper bound. Will 2017 do the same? We will see shortly. I have not worked up daily data for stations further north, so I cannot run the same absolute comparisons. Note of warning: August NAO values have turned strongly negative, which is generally where they have increasingly been since about 2005. Note the last NAO chart at bottom at his post. We have not experienced a strong negative NAO on the declining flank of a solar cycle since solar cycle 19. Note the 1962-63 NAO placement.
|
|
|
Post by nautonnier on Sept 15, 2017 1:13:59 GMT
|
|
|
Post by Ratty on Sept 15, 2017 7:11:25 GMT
What's the worst that could happen? Don't answer that ......
|
|
|
Post by blustnmtn on Sept 15, 2017 11:38:50 GMT
1938...before history when dinosaurs roamed the earth.
|
|
|
Post by acidohm on Sept 17, 2017 19:43:15 GMT
|
|