Side note: the “forecaster’s discussion” for your local NWS forecast area is easily the best one you can find. It’s the one all your weather apps and local TV meteorologists crib from. Sure, there’s a bit of meteorological terminology to learn, but they often link off to a glossary (then you can put that term into Wikipedia).
My favorite part of the forecasts is that they tell you unequivocally when they are confident in something and when they are not sure. They are not afraid to say a qualified “we don’t know”. This is is a stark contrast to almost every other forecast falsely projecting complete confidence trying to vie for viewership to attract advertising dollars.
I keep seeing people talking about how X Weather App is better than Y Weather App; I've been using a simple bookmark to my local forecast.weather.gov page for years and it has given me an accurate forecast in plain HTML every time. Plus there's radar and satellite imagery. It's great!
I know that this is a little off topic, but speaking of X Weather app.... The habit that I am still trying to break with my kid is that they should not rely on the weather app for temperatures when we go camping.
The weather app is an app that reports data from a weather station, it is not a thermometer. If you're camping in a valley or a mountain that's 20 miles away from the nearest weather station, the temperature that your app says is almost certainly wrong.
So hard to break that habit of relying on the phone when it's not correct with some of the younger folks.
In the days of Gopher and WAIS, they would talk about the different supercomputer models being in agreement or diverging. The NAM would say one thing but the ETA would say something else. If you remember what ETA was, good on you. :)
Someone replied, but deleted, asking about forecasting and their values and I'm wondering the same. Is the weather forecast an expected value, like say an average given a forecasted upper and lower bounds?
They are running ensembles of models to predict the forecast. Chaotic systems are challenging to model, and models aren't perfect, so a collection of models is used (either different initial conditions or different versions of the model, as per https://forecast.weather.gov/glossary.php?word=Ensemble).
This gives you a distribution of predictions, which you can use to produce an average, as well as uncertainty, confidence bands, etc.
The quality of these predictions and ensembles is empirically validated using proper scoring rules (https://en.wikipedia.org/wiki/Scoring_rule), which is a robust way to measure the quality of a probabilistic forecast. Negative-log-likelihood, a ubiquitous loss function in machine learning, is an example of a proper scoring rule (with a minus in front of it).
The science behind forecasting has gotten pretty sophisticated, it's an interesting field.
I once heard that precipitation forecasts represented the amount of land in the area that would see rain, not the probability of rain for the total area. I.e. 50% meant half of the area would get rained on, translating to a 50/50 "chance" for you.
Then someone else told me that was bogus, and I suspected as much.
Chance of rain = (chance you will be somewhere in the forecast area where it is raining) x (chance it will rain at all) [1]
The two are independent variables, the overall probability of two independent events both occurring is the product of either event occurring. The chance of rain is the product of two independent events.
So, 50% could mean a 50% chance it will rain everywhere, or a 100% chance it will rain in 50% of the forecast area, or it could be anything in between. (Eg, a 1 mile square cloud that is dropping rain travelling through a 10 square mile area, or a 10% chance that a rain forming cloud will develop over an entire area, or anything in between is a 10% chance of rain. The first case can often be observed with radar and visually, an hourly forecast can be very accurate to know that one cloud will still be raining in an hour. Hence the size of rain area and size of forecast area are variables that come into play)
As a few counterpoints:
- Multiple models do not mean each model is equally likely.
Nor do multiple models mean we cover the set of all possibilities. For chance of rain to be the fraction of models predicting rain, those two statements would have to be true
- models themselves each give a chance of rain. These different values are not combined in any way, forecasters pick a model
Some implications and interesting points about forecast area:
- if a forecast area is "zoomed" out, should that alter the chance of rain? [Yes, this implies that chance of rain has forecast area as a dependent variable]. Hence, if you "zoom in" on a forecast area far enough, the chance of rain developing becomes the dominant variable for chance of rain.
One interesting thing is a forecast assumes equal probability that you will be at any given location within that forecast area. Which means the percent area forecasted to receive rain is the same probability that you will be at any given location within a forecast area receiving rain
I'd like a source for temporal correlations. Their models definitely have enough data to differentiate between the following two hypothetical predictions (albeit, predictions are chaotic, but the ability to predict this sort of thing is no worse than the other predictions), but I have no way to parse the result to obtain them:
Suppose the hourly forecast has 50% chance of precipitation each hour for two hours. In addition to the points you raise about how to interpret 50% at any moment in time, how are we supposed to know if this means either of the following (or something else):
1. A big storm is passing near a point for those 2 hours, and the only question is whether it passes over it or to the side. I.e., both hours have rain or both don't.
2. A small storm is definitely passing over that point, and the only question is when it will arrive. You either have rain in the first hour or the second but not both or neither.
More like the whole chimp troupe will agree that chimps can't control the path of a storm. With such problems there is no pressure on the forecaster.
But the moment one chimp stands up and says no no I can, and a large enough group is persuaded(even if its false), the game changes. Forcaster will start behaving differently. They aren't machines. They are chimps.
"""
To put it simply, this will likely be one of the most impactful
systems on a widespread scale that this meteorologist has seen in
a long while. The impacts will include widespread flooding, roads
washing out, hillside collapsing, trees down (potentially full
groves), widespread power outages, immediate disruption to
commerce, and the worst of all, likely loss of human life. This
is truly a brutal system that we are looking at and needs to be
taken seriously.
"""
It will be 3 inches of rain on fully saturated soils. This will lead to 3 inches of pure runoff on a landscape that rarely experiences such runoff.
So hillsides, channels, creeks, rivers, and so forth will all be experiencing record run-off levels. This is why the forecast calls for significant hillside failures: it will be more water on a hillside than many of those hillsides can bear. It will flood beyond the normal flooding limits. It might... be historic. Maybe. Huge if. But super-dangerous if it happens.
3 inches in the Olympic Peninsula would be less of a big deal because that's routine. All the hillsides that can fail with that already have. So you're right that 3 inches is not a big deal in most places. But on landscape where that doesn't ever receive that much, it suddenly is a big deal.
The fear I have with these recent atmospheric rivers is that we have a lot of snow in the Sierras now. If we get an atmospheric river which is too warm, bad things could happen[1]. Fortunately as we get deeper into Winter it's probably too late for that disaster to be possible.
>it will be more water on a hillside than many of those hillsides can bear
A good example was the Oso mudslide. Nearby Darrington gets 80 inches a year. (Twice as much as Seattle!) March 2014, a storm drops 3 to 5 inches on soil saturated by six months of rain, and a hill lets go. Killed 43 people: https://en.wikipedia.org/wiki/2014_Oso_mudslide
I've driven past Oso many times in the last few years—it's a solemn drive. It's also a tiny reminder of the planet's ability to shrug its shoulders and wipe out civilizations.
That's about what the South Bay saw on the NYE storm. Peninsula & SF got about 5", mountains up to 12". Seems to be roughly comparable in raw precipitation.
The concern over this next storm is that the ground is already saturated and the creeks are already overflowing. In a normal rainstorm like the one we just had, the first inch or so is absorbed into the ground, then the rivers start rising and mudslides start happening. This time, we're going to start seeing those effects from the very start of the rain (which is already happening - it's raining now).
To put that in context, snow/water ratios are normal about 10:1, so if what was snowing you’d expect close to a 1.5 feet to 3 feet of snow- that’s a lot, as my snow shoveling back is telling me.
If my quick internet search is correct, SF will be getting well over 10% of its average annual rainfall in just a day or two. Another way of saying a lot.
Also, the water doesn’t just stay there, it flows, and it makes soil looser and heavier. 3” in the hills can mean many feet of extra water in valleys/flood plains and can cause mud slides.
My recollection is the SFBA gets around 20 inches of rain annually. Three inches in a few days time is 15 percent of the rain it is supposed to get all year. It's a lot. The landscape won't cope well, especially if it's already seen rain prior.
We saw 2 inches of rain one day in The High Desert, an area that gets 6 inches annually. The flood waters took weeks to recede and brine shrimp hatched out.
You probably just saved my life. I was planning to drive the coast highway on Wednesday and Thursday, including the Golden Gate Bridge and Monterey Bay... Time to replan this trip...
Be careful, today a tesla fell off cliff at Devil's slide which is between Santa Cruz and SF. Stay away from major freeways as well: slides are expected. Central valley freeways may get flooded.
That had nothing to do with the weather. Tesla makes very heavy cars, some of which have brakes that don't match the performance implied by their acceleration.
Meanwhile Devil's Slide is in an area suited for low speeds, careful acceleration, and reliable braking.
With witnesses are describing this:
> "As I’m driving by, I’m like, ‘Wow, he's driving extremely fast to take that exit. You're not even supposed to be going up that way,” she said. "And I can see in my rear-view mirror this car that just go over the edge and straight down."
It's hard to imagine it was anything but driver error.
Devils slide is bypassed by a tunnel. I remember before the tunnel was built riding in a car with my brother in law and he just floored it through that super narrow section practically hanging off the cliff over the water. I asked him what he was thinking and he said he didn't even notice.
Mountain passes across Sierra will be closed, so you can't travel to Reno, as I-80, SR-50 (Sacramento to south lake Tahoe), SR-88 (Carson Pass) will be closed. Even Altamont pass/I-580 had a slide a few days ago. SR-99 will get flooded, as it is parallel to I-5 from Sacramento to Bakersfield. So, expect long delays on both I-5 and I-80, since other secondary roads get flooded.
If you are planning to Vegas, you gonna have problems on SR-152/I-580 to I-5. Also, expect problems on SR-58 between Bakersfield and Barstow on I-15. Even if you want to go to Sacramento, expect long delays on all routes.
Whichever freeway you pick, you will have problems: bad visibility, very slow traffic, partial closures due to puddles, complete closures due to floods (like the north of SFO airport last time).
That's why CHP, Caltrans, NWS advise people to postpone their travel unless extremely necessary. No storm tourism either.
Carson and the other smaller passes close in the winter and open sometime in spring. This happens all the time.
I suspect 80 will be closed maybe overnight because way too many people think driving in snow is just like a dry road - spin outs and crashes (even semis). In general CalTrans does a good job keeping it mostly open.
For context - I live north of Truckee and deal with 80 regularly.
Carson pass (88) is plowed and stays open all winter. It's the road to Kirkwood and connects to South Lake Tahoe via 89. Agree all the other smaller Sierra passes are closed after first significant snowfall.
It's probably time to hunker down Weds/Thurs if you're in California.
Also, a lot of the small mountain roads between 1 and 280/101 washed out in the New Years Eve storm, so if you are stuck on the coast there's no guarantee you'll be able to get inland.
A bunch of the fairly major urban (what passes for urban here, anyway) roads were in bad shape on NYE. 101 had partial and complete closures in places due to flooding. Sections of road that are local elevation minima (e.g. under the Caltrain tracks) were just standing water anywhere that drains weren’t working well enough.
It's no fun to drive in the rain, so either pick a nice place to stay for Wed/Thu during the storm, or postpone the trip. If you need to get south, just drive the 5, it shouldn't enough rain to flood it. The central valley flood alerts you are seeing are for rivers and small roads with stream crossings. Forecast there is only 1.5".
Parts of 99 were closed (with at least one fatality) and I-5 threatend by the flooding this weekend; the next storm is bigger and will hit ground that is already saturated. The flood watch issued in the Valley for the upcoming storm is not just for rivers and small roads with stream crossings as you describe.
> Forecast there is only 1.5".
The Centrall Valley forecast for the storm is 2-3”, actually, per NWS Sacramento.
Thanks for all the replies. The trick is: I'm halfway through my trip right now. So I think my choices are 1) stay where I am until Friday, or 2) back-track to Oregon and head for I-15. Seems like staying here is my best bet. But I guess I won't be at work on Monday...
Looks like things will be getting ugly Wednesday afternoon.
How about hitting the road before that to be ahead of the mess?
Monterey and Big Sur are great, but just getting home seems like a better idea.
Even if you can't leave Tuesday, you did say you had planned for Monterey on Wed. So leave early Wed, cruise down I-5 and you'll be at the grapevine with time to spare.
If "I-15" is shorthand for "395 to I-15", I would recommend against it. Cold snow after wet snow is a manor avalanche condition. It would be unlikely to hit you of course, but the roads may become impassable in ways that are far worse than I-5.
Forecast for the Bay Area & Monterey has the storm starting to roll in at midnight tonight, so there's relatively little time to get out if you're planning a long road trip out of the storm's influence. Unless OP is hitting the road right about now I would recommend staying put.
New Year's morning I drove along the coastal highway for a pretty good stretch, and the conditions are already a little sketchy, but I don't think it's so dire that you'd just get washed away or something.
The biggest risks I encountered were partial flooding in low lying sections that bank inland, and various sized rocks. No signs of new erosion along the edges of any sections I encountered.
Edit: to be clear though, I wouldn't say this is a time to take a joyride in the area. I'm more speaking to the fact it's a bad idea for a fun vacation right now, not a death sentence
We've had some pineapple express systems come through the PNW and they were unbelievable rain events. One completely flooded the apartment complex bottom units where I lived.
Good luck bay area folks, definitely take it seriously. If you're a home owner and have been putting off cleaning out gutters, drains, etc. absolutely get on it right now while you have some time. These unusual rain events will easily back up semi-clogged or poor flowing drainage systems.
My garden level apartment was destroyed by flooding on December 23 of 2020. Homeless in Seattle with a wife, a kid, and two dogs and two cats for Christmas.
To this day I have ptsd when it rains hard, particularly around Christmas. Now I live on the top of a very big hill…
I got flooded in a 1st floor apartment once. After that my apartments were alway 2nd floor and up. At least until i became a home owner and then I made sure my house was on well drained gentle sloping area. My real estate agent thought I was being overly picky, but fool me once...
I put a bunch of USGS river gauge data into a notebook on Saturday and found that Bay Area creek flows were either above or near all-time records. Arroyo Valle, Alameda Creek, San Francisquito Creek and many others set records. And they’re predicting this next storm to be bigger and longer.
For us the answer was just "get you and your loved ones to high ground, forget about the dang servers, let them go down, it's just computers". Refreshing. In any case the correct investment is in automated recovery systems and a business plan/partnerships that don't demand unrealistic SLAs.
Nah. Part of the narrative was that various Availability Zones (AZ) in a single region would be on "different flood plains", and be attached to different power grids.
The essence of this design is that a single event, affecting either the ground (floods) or electricity (a grid), would not be able to single-handedly kill all AZs in an AWS region.
Source: I used to share this stuff a lot, in public, representing AWS.
The part about how entire groves of trees could be toppled is wild to think about. I imagine (though don't know) the supersaturated soil means their footings have turned goopy and a strong gust could just push them past their holding limit.
Last week here in Oregon, a storm with 50mph gusts came through and blew over two groves of trees here at my farm. Dozens of trees crashing down in the space of hours. Got very lucky in that they all managed to avoid doing any serious damage to structures or vehicles.
My first floor flooded on NYE. I'm not looking forward to Wednesday -- will likely flood again. Going to be a lot of $$$ to repair and fix the drainage.
See: Hurricane Floyd. It was so bad because ANOTHER hurricane had dropped 2-3ft of rain on Eastern North Carolina a few weeks earlier. As a result basically everything east of I95 saw a 500 year flood.
It would be super great if CA had invested in water recovery and reclamation to any real extent in the last few decades - because there's been little-to-no investment in the infrastructure not only do we run the risk of infrastructure collapse (a la Orville) but it doesn't even really do much to help with the water undersupply once the storm is gone :-/
Maintaining the existing infrastructure - I would have thought "duh", but Oroville demonstrated not - but there's also various river outflow and levee systems to stop water immediately washing out to the sea without being slowed to allow seeping into the ground reservoirs.
Oroville was actually rebuilt and held up nicely in recent atmospheric rivers. Same with the South Bay dams that caused the San Jose flooding in 2017. I wouldn't use those as examples - if anything, they are examples of how California actually can build infrastructure, quickly and reliably, when there is profound media embarrassment and multi-million-dollar lawsuits at stake.
The infrastructure you have to worry about is the stuff that everybody's forgotten about, eg. the levee and dam systems in the Delta.
Oriville dam is infrastructure that was forgotten about. We only know about it in any immediate sense now because it collapsed due to the failure to maintain it. Which was my point.
Seriously, any “cost cutting” measures that “defer” maintenance of infrastructure should be required to put aside funds to rebuild that infrastructure.
Oroville dam did not collapse. The main spillway developed a large crater in it that put it out of operation, and there was serious worry about the emergency spillway collapsing, but the dam itself was never in any serious danger of collapsing. (If it had, you would've heard about it. The impact of the emergency spillway collapsing would've been a 30 foot wall of water going down the Feather River; the impact of the dam itself collapsing would've been a 700 foot wall of water going down the Feather River.)
Both the emergency and main spillways were fully repaired in 2018, and the repairs have held up nicely in subsequent rainstorms. You probably didn't hear about the repairs because the media never reports on stuff that goes as planned. There are several case studies about it on civil engineering blogs & channels though:
The reduction of flood plains, and the removal of natural dams (beavers!) means that what you mostly get is high speed run off that is high in sediment and low in oxygen.
Up and down the Bay Area Peninsula e.g. SF, Hillsborough, Atherton, Redwood City, Menlo Park, Palo Alto, Lost Altos, Mountain View, Sunnyvale, Cupertino, San Jose, Woodside, Saratoga, and so on. As well as plenty of other areas around the bay.
Surprising amounts of Chinese investor houses and just plain vacant homes. Some forever vacant due to multi-year ever expanding build and remodel projects.
People laundering cash acquired from dubious deals done outside the country likely don't care about making a little more money - the principal aim is to keep the cash in a hard asset that can't be seized by the local (to the purchaser) government.
I doubt it happens a lot but there was an unoccupied house in my city that was burned down by an escaping suspect. Turns out it was owned by a lady living in China who invested for a future residence. Given that the Chinese real estate market is overheated it makes sense to diversify.
Secret Samsung Executive Party Houses, vacant investor houses, I've seen it all firsthand. I don't have an issue with it except in cases where it takes desirable housing out of circulation away from people who actually would like to live there.
My perception is the Atherton and Hillsborough NIMBYs cause far larger waste and trouble than the Foreign Investor homes who participate in the existing systems. The NIMBYs are the real xenophobes manipulating local political climate, city ordinances, and zoning preventing new housing from being developed.
For the downvotes, less than 10% of transactions in the US in 2013 [1]. When I first saw this for Canada I looked it up and it really only accounts for something like 3% there.
When that small percentage is concentrated in the most desirable areas it becomes amplified and adds significant strain to a housing system which wasn't designed for it.
The really interesting data would be the geographic spread. 10% of all total transactions in the United States is actually a huge volume and big deal when concentrated in NYC, SF Bay Area, and Seattle etc.
If you keep predicting quakes, one day you will be right :)
I've lived in the Bay Area since 1995, and always wondered what would happen if a substantial earthquake struck while the hills are already half mud from sustained winter storms.
The article said that prediction is by a quack site, with no scientific background, who missed both the real Eureka earthquake and then falsely predicted a magnitude 6.5 in Eureka on Dec 25-28.
In the South Bay Area several of the creek trails and adjacent parks have been taken over as homeless camps. In particular the Guadalupe River near downtown San Jose and SJC. They're going to have a rough time.
When I first moved to SJ I tried to take a walk along about 5 different bike trails, along different creeks. I made it a few hundred feet along each, before turning a corner into an encampment and getting nasty looks from the residence. A near jab of a needle on the ground ended my desire to try any more. Some of the encampments were really massive. I had no idea.
Saratoga ended up having the only family friendly creeks that I could find.
I don’t understand why anyone would downvote your comment, except for denial of the reality that has become the Bay Area.
This title and the weather report make me wish people with factual jobs would stick to the facts.
It's really not a good sign when factual government projections start getting jazzed up with personal history.
All things considered, this storm is well within the design considerations for any road or building in California, so it shouldn't be considered, in advance, as catastrophic.
Is there no where to avoid relentless self-absorbed posturing?
Disagree. NOAA weather reports are basically perfect. You know shit's getting real when the report isn't boring to read. They do a great job of putting little personal touches on them, without letting it get out of hand or get in the way of providing information—which is precisely what professional writing ought to look like, in many cases. This makes them far more useful to a layperson than a bland recitation of facts would be, since it provides context that might otherwise be absent or harder for a non-expert to appreciate. When it comes time for actual warnings they keep them to-the-point but also don't pull punches or dance around the seriousness of their assessments—nor do they hype them without cause, generally.
I've wondered for a while how NOAA reports are allowed to be so good when you'd think the march of bureaucracy and bureaucracy-fans and liability/anti-responsibility pants-wetters taking the humanity and personality (except wholly fake and brand-conscious "personality") out of everything would have gotten to them by now. Please don't encourage that :-(
My favorite part of the forecasts is that they tell you unequivocally when they are confident in something and when they are not sure. They are not afraid to say a qualified “we don’t know”. This is is a stark contrast to almost every other forecast falsely projecting complete confidence trying to vie for viewership to attract advertising dollars.