Sub/Alpine meadows, fires, & climate change

[retired jerry]

well said

and, I've read that if temperatures are warmer, then water evaporates more which will make the effect of the drought worse.

I got those plots from that cliplot link you posted. Lots of good data there.

It's funny when someone makes an (intentional) pun and says "no pun intended". "But your pun was intentional, why did you say it was not intended???"

[drm]

To make one point a bit more quickly, there is meteorological drought - how much it rains (or doesn't) and ecological drought - how dry the ground is. And the latter is what matters for the most part, especially if discussing meadow survival. The latter takes other factors besides rain in, including high temperatures that increase evaporation. I read that increased temperatures have effectively lessened the soil moisture equal to a 10% drop in precipitation, but that was for a broad area.

I think there is nothing safer to say than there are multiple factors driving the rate of meadow fill-in. And I wouldn't even assume that the factors, whatever they are, are the same across all of Indian Heaven. I could easily see difference intensities between Indian Racetrack and the meadows between Junction and Lemei Lake. Soils, drainage patters, exposure to sunlight - all of these can vary.

Lastly, I have noticed a trend that summer conditions overwhelm winter and early spring conditions. I follow Snotels pretty closely every year and have seen many years when the snow level in Feb/Mar/Apr was above the 30 year average and yet the snotels still hit zero snow earlier than the 30 year average because we had a warm late spring - though obviously not this year. This means a longer dry season. And although that didn't happen this year, we had strong drying conditions will into mid-October. It wasn't raining yet AND it was very warm.

I bet the glaciers on our volcanoes took a big mass ice hit this year too, like many recent years.

[BigBear]

Jerry: you inadvertently proved my point

The Oregonian used to include the rain bucket for both the calendar year and water year in their weather page, and then stopped doing that when "drought" and "above average" rainfall didn't coincide. Sure, there's much discussion about how one determines if its a drought year, but it's really hard to make that determination when the numbers aren't there to fact check.

Many years before the pandemic i used to track the daily high/low, and I stopped doing that when I needed to update my chart after being on vacation and I went to Accuweather which used to have this information at that time. I found a rather alarming situation: the temperatures for the last week I had entered the data in real time had changed, so had the previous month, and all of the months previous. The numbers were different by 1 or 2 degrees on virtually every day. I asked one on my hiking friends who was a meteorologist and he said: yes, we periodically update the numbers to reflect the model. I asked, weren't these numbers actual readings? He said, yes, but the model is more accurate. As an accountant, that made my head spin: the model is more accurate than the actual readings.

But the fun of number manipulation is in real time too. A couple of weeks ago, I was watching GMA and they were talking about the devastating hurricane that hit Florida: it was one of the ten worst of the century, no, 30 seconds later, it was a 50-year event, and the next novice reporter said it was a 100-year event. I kid you not, ABC national news that night: "some are calling it a 100-year-event." Who are the sum, their own novice reported who pulled the number out of thin air. That's how fast the facts could spin.

So, let's move onto Jerry's proof. Real time numbers verses 4-month reflections. I sometimes keep weather notes in my hiking summaries. And in early July, this was a news story: From January to June, Portland received 23.43 inches of rain at the Portland International Airport, well above the average of 19.68 inches. But that’s only the 16th wettest first six months on record since 1939. The wettest first six months was in 1996 with 30.86 inches. So, here we are at the end of October and that "wet" period became, what was it, average or slightly below average? No, I'm afraid I didn't write down which day in May the rain year bucket got filled, but it has evidently evaporated since then.

Basically, we are living the life of Animal Farm. The paint on the barn wall is dripping and viewers have no idea. Of course it's rather hard to prove unless you methodically track the information in real time and then compare it to the newest spin. It's something you would have to have the energy to do yourself, which no on does, especially after you realize they are always updating and your whole effort was merely academic.

Oh, the other new item this week: Ginger Z on the glaciers in Switzerland. The noted comment: they haven't melted this fast for 2,000 years. So, the increase in temperature isn't necessarily accelerating with time. It has periods of flux. And, as I mentioned some time ago, it predated fossil fuels by thousands of years.

[retired jerry]

you would really have to have your own rain gauge

yeah, we are living in an Orwell world for sure

up is down and down is up

was that Orwell?

[retired jerry]

one more plot and then I'll stop

Klamath Falls - 53% of normal.


stories I've heard in the news are that as you go south and east, like to Klamath Falls, the drought is worse.

this is consistent with these plots

this is consistent with stories about not having any water for irrigation

maybe they initially release data, then later update it when they get better estimates. That doesn't mean they're lying about the data. Besdies, if they wanted to lie about it, they could just lie in the initial data.

[Webfoot]

Besdies, if they wanted to lie about it, they could just lie in the initial data.

It would be more noticeable if people are thinking "I was there and that's BS" the day they report it. But besides obsessive collators who will notice the change six moths later? The strange updating to "fit the model" that BigBear reports should not happen for something as simple as a temperature or rain gauge. It's not something that requires convoluted estimation like total snowpack in metric tons.

More fun with flags charts.

[retired jerry]

wow, a lot of precip in May and early June, the drought seems to be more south and central. Large areas in California.

driving over Shasta Reservoir on I5 is always interesting. Or Trinity Lake. There is definitely drought going on there.

no drought in Indian Heaven though so it will be overtaken by trees

[drm]

Many skeptics of actual temperature readings have complained that location and other issues result in exaggerated readings due to urban heat island effects and the like. I suppose models can take into account readings at a number of locations locally, and so average out factors like urban heating or just funky monitors. And in fact, I did a search on "actual temperature replaced by models" and found that somebody who compared the two found that model results were slightly lower.

It really depends on what you want to measure. Do you want to measure the temperature at a very specific location, or do you want the temperature for that city or region? The habit of using airports for the official temperature of a city seems

[sgyoung]

It really is common practice to weigh individual results when aggregating across many heterogeneous sources. This is done in survey data all the time to account e.g. oversampling of a certain demographic, etc.

It doesn't seem weird or fishy at all that a similar process happens when collecting meteorological data. These methodological practices are standard and transparently reported. It's really annoying to see it implied that this is a bad practice or done to cook the books.

In fact, feel free to google "weighted climate models" and enjoy page after page of results that demonstrate how this is done (get ready for some stats) and why. One the benefits of weighting is that it allows you to correct for systematically biased observations from specific locations.

Can I ask BigBear how they interpret the changed data mentioned in their post? Do you think it is done dishonestly to push a narrative? Do you think it's bad practice? And what evidence do you have to support your concerns

[BigBear]

sgyoung: I wouldn't go as far as saying it was done with malice, I think it's more likely the result of wanting the model to be more accurate than it really is. It's never honest to change the actual numbers to match your predictions, but it is happening. I do find it interesting that when talking about Portland's rainfall because there is only one official number. There should not be any reinterpretation.

One example of poor statistical analysis (not related to weather), but one which I am well aware of because I was one of the participants of this Harvard study, was a subsidized housing questionaire which asked questions about costs HUD never analyzed. So, respondents which answered "unknown" or left the space blank were assumed by Harvard to have $0 cost. As a Master student of Quantitative Analysis, I know that such an assumption makes the model invalid - you are required to ask for further input or create a new sample and start over. Just assuming "unknown" equals zero is as incorrect as changing the actuals to accommodate the model. It's not necessarily malice, but it is laziness. In this case, Harvard was clearly relying on its name. There was quite a bit of feedback on this issue because the majority of the respondents left many spaces blank or unknown.

The other issue with model development is not properly analyzing whether a condition actually effects a model's predictions or is just an existing condition with no true relationship, resulting in correlation coefficients. In this blog, the issue is what is causing global warming since it has been happening for over 10,000 years. Population has been increasing, forests have been burnt to raise crops and urban growth, wood fires across the planet are more common, and yes, the burning of fossil fuels has increased. But one of these issues has caused global warming. They aren't helping the situation, but if human activities ceased, the world would continue to warm.

Another interesting issue which you can test out right at home right now... There's a lot of discussion about oceans rising because of the melting of ice sheets. It's said like it's a scientific fact. I will agree that the melting of ice on landmasses could increase the ocean's level, but most ice is floating in the water...and the melting of that ice will have a zero impact on the ocean's rising or lowering. So, how can you test this? Remember high school? Fill a glass with ice and top it off with water. Watch the ice melt. There may be a ring of water at the base of the glass over time, but that water is coming out of the air and condensates on the outside of the cold glass. However, the ice in the glass interestingly will shrink as it melts and thus maintain the same water level in the glass.

The water on land which enters the oceans will make the oceans rise if it is not offset by the rainfall occurring over the land. An interesting dilemma now arises: if dams hold back the water and you are truly concerned with oceans are rising, shouldn't you create more dams? The single largest "sue" of water is the natural runoff to the oceans. It's something like 95% of rainfall makes this journey without being used as irrigation, manufacturing or residential.