According to the Water Emergency page on the Albemarle County Service Authority’s website, reservoir supplies have increased from 53.1.% of capacity to 53.9% of capacity. Given that an average of 1.5 inches of rain fell across the area yesterday, we can calculate that a 1% gain in the reservoirs requires approximately 1.9 inches of rain. Therefore, we would require just over 30.5 inches of rain in order to reach 70% capacity, at which point restrictions would be cut back. Good news from ACSA, as reported in the Progress, is that area water usage has dropped from an August average of 13.2M per day, to 10.4M throughout most of September, and is now down to 7.8M as a result of the most recent water restrictions.

Very glad you are keeping track of these things. I would love to see a daily posting of current reservoir supply and current water usage rate.

But–your math is based on the assumption that x inches of rain equals y more in the reservoir. It is in fact a very complex calculus, figuring in how fast the rain falls, relative soil moisture, how much the ground can absorb and how much runs off, how dry the air mass is and hence how much evaporates, &c.

Let me simplify things and say: welcome as it was, that wasn’t enough water to fix the drought. We still need to conserve.

This page offers a daily update on the reservoir level (thanks to Jinkster for posting this link on a different thread). I’m not sure where one can find a daily update on the water usage rate, though.

But–your math is based on the assumption that x inches of rain equals y more in the reservoir. It is in fact a very complex calculus, figuring in how fast the rain falls, relative soil moisture, how much the ground can absorb and how much runs off, how dry the air mass is and hence how much evaporates, &c.Absolutely. But since neither the city nor the county appear to be offering any solid roadmap or even history of this sort of data, my math will have to suffice until somebody smarter comes along. :)

There is a reason the National Weather Service is one of the main markets for supercomputers. There are so many itsy bitsy changing numbers to keep track of, that only a machine can do it. The flap of a butterfly wing in Africa is a hurricane in Louisiana. That sort of complexity.

As for me I’d rather go outside and look at the vegetable garden. From this I learn: there’s a bad drought.

As for me I’d rather go outside and look at the vegetable garden. From this I learn: there’s a bad drought.:) Too bad more people don’t think like that!

Sure it’s complex and Waldo’s calculations leave out a lot of things. For example an aquifer that feeds a spring that eventually helps fill the reservoir may actually be recharged by rainfall from miles away and outside the reservoirs watershed. Even so, Waldo’s calculations will help people understand that this is a long term problem that is not going away anytime soon and that it’s going to take a lot of rain before the drought is over. I put it up on a bulletin board at work.Thanks Waldo.

BTW, I hope that anybody will feel free to refine my hasty calculations with any other data or improved mathematical processes. Those figures are as quick and dirty as they look. :)

Waldo’s math is fine with me. The stat I’m worried about is this (from the Water Emergency page):

Do you think people saw cloudy skies and felt they could go back to taking long showers again?

My appologies in advance to all my math teachers.

Waldo’s calulations are perfectly valid, and not that much different than the ones the city experts quoted. Both assume a linear function of supply and demand.

The most accurate tool for modeling complex systems where we have a small statistical sample, but no way to observe the complete system is currently "bayesian probability theory" which is just a big word for an equation that models uncertainty by combining common sense and observations. Obviously the color of my shirt has no effect whatsoever on the resevoir level, so we don’t model the entire planet to find this out, we just identify the variables that might actually be usefull. Rainfall, area, slope, humidity, air temp, etc.

Once we have our variables, we need to decide how they are connected, rainfall and area determine volume, volume and slope determine flow rate, humidity relates to loss in the system.

This is where it gets really interesting, We can either make assumptions, like "all water will make its way from raindrop to reservoir in 12 hours, and loose no water to vapor". Or more accurately, we could look at observations and say "when we get one inch of rain, the lake level goes up X percent." Thats exactly what Waldo did, allbeit with an incredibly small sample time. Perfectly legitimate starting point.

The power of Bayesian modelling is that it calculates probabilities for each connection between the variables of interest. We can pin down one variable, and run the model again and see how that effects the outcome. Of course, this requires extensive computing, but today’s computers are up to the task. Bayes theorem was proposed in 1761 by Thomas Bayes and remaind a curiousity because only trivial cases could be calculated until the late 1990’s. Recursion baby! Oh yeah!

Explaining the mathmatics is pointless. Aside from the lack of mathmatical symbols in our character set and the limited time we have available, it really would hurt your brain far too much.

Most college educated folk would assume its a calculus. But in reality, sane people calculate discrete states. For a river or spring, it would be declining, no change, or increasing flow. 3 states. If we segment the time series down to sets of discrete states and state changes, we can predict the "hidden" state changes with the observable time-segment. In this case, the hidden state changes indicate flow rate.

This is a type of stochastic approach called a "Hidden Markov Model". Consisting of a "Markov Chain" of segments, each of which have a deterministic probability of being any given state. These numbers feed into a maximum liklihood or maximization function.

And then this number pops out, and its still wrong, but less wrong than Waldo’s number. But I’ve become fond of Waldo’s number, so I’m not going to bother calculating it. And I seriously doubt that anybody else will be motivated to either.

Bayesian probability is also known as subjective probability, personal probability, or epistemic probability.

Like all imperfect theorems, weighted basis factors are applied. And guess who determines both the factors and the weights?

I have no problem with Bayes. I have a problem with the resulting school of thought that entrenched in almost every facet of our lives today. In essence, the Western world, and the U.S. in particular, believes not only that everything is identifiable (factors) and quantifiable (weights), but that the most complex subjects are already mastered.

As long as Bayesian Probabilities are considered "educated guesses", I’m fine with them. But when they become dogmas, as in "U.S. capitalism is the only viable socio-economic solution", then that’s where I get uncomfortable.

Back to the water levels, I think silkyzephyr gave a good enough descriptions of the major factors involved.

I agree.

This is very true. There’s also the question of ground water, which feeds mountain springs, which in turn feeds reserviors. There could be no rain for weeks, but if the underground aquifers (sp?) have a lot of water (which obviously they don’t right now), the springs will still be running, which would help to maintain the reserviors. Lately of course, the ground water is terribly low too, so less spring water comes out of the mountains, so less water goes into the system.

So rain helps in the short term by flowing directly into the water supply, but also in the long term by keeping the underground water table higher, which benefits everyone.

I’d also like to offer a minor quibble with the notion: "water usage . . . is now down to 7.8M as a result of the most recent water restrictions."

No the drop is not because of water restrictions. It is because of good citizens like Waldo and many others who no longer shower every day, don’t run the sink, don’t flush as often. Little sacrifices which add up to big savings.

We are perhaps too quick to give government credit for what citizens–individual citizens–do. For most of us, the restrictions are beside the point. We have no carwash to close or lawn not to water. We citizens are saving water on our own, each of us as individuals, because it is the right thing to do.

I’d also like to offer a minor quibble with the notion: "water usage . . . is now down to 7.8M as a result of the most recent water restrictions."No the drop is not because of water restrictions. It is because of good citizens like Waldo and many others who no longer shower every day, don’t run the sink, don’t flush as often. Little sacrifices which add up to big savings.I certainly hope that you’re right. It’s too bad that we haven’t been given data — should it exist — regarding the source of the savings. How many gallons are saved each day by shutting down all the car washes, by ceasing to water lawns, by restricting the watering of plants? How many, after that, are left unaccounted for, indicating that it’s citizens’ voluntary efforts? My estimation skills are notoriously worthless, but I’m inclined to side with you in the conclusion that the bulk of water savings are not a result of the new water restrictions, but a result of individual conservation efforts beyond the legal requirements.

Notice I said the best method CURRENTLY is bayesian analysis. And I also said it pops out a number that is wrong. I agree with your dogma dogma :) There are 3 kinds of lies. Lies, damn lies, and statistics.

It is however incredibly powerfull for problems such as protein structure prediction from sequence. Something where you have the protein structure (determined with x-ray crystalography) and the original gene sequence of some proteins, but only the gene sequence of others.

Any problem where you have a choice between a 98 percent chance of being right or not knowing at all is a great use for such statistical nonsense.

I didn’t explain all that so that we could calculate it. I explained it for those who are curious how the "experts" would go about attacking the problem. And still be ALMOST as wrong as you or I.

Certainly its a lot of calculations, requiring significant computing power, but physical modeling of the actual system would be impossible. So its our only choice.

:)

Does anyone here work for the water authority?

What is up with that figure? How does the equipment work? Does a woodpecker read the daily flow off a gauge and carve it into a stone tablet?

An hourly or better yet 5 minute average would be sweet! We could point to it and say "people who are lazy and wake up at noon shower longer than people who get up at 6"

Better yet, open the utility billing database and let us hunt down and taunt the worst offenders into reducing usage.

Also, is there a discrepency between the billed amount, and the amount of water leaving the facility? You know, like the transmission line losses in an electrical grid. We could be loosing water out of leaky 100 year old pipes and not even know it.

Well, since there is NO WAY IN HELL we are going to get 30 inches of rain, it’s pretty much a done deal that we’re pretty much screwed.

Since there is no hope, let’s forgo the doom and gloom and just carry on smartly until the tap runs dry.

Jinkster’s post here offers pretty strong circumstantial evidence: "Thursday’s demand was 8.5 million gallons, up from 7.75 million on Wednesday."

That was not because scofflaws rushed out to hose down their lawns during a rainstorm. Nor did a skulking parade of cars line up to squander water at illegally re-opened carwashes.

The rise in water use happened because as welcome rain pattered on the roof, people relaxed their vigilence a little: "at last I can do the laundry and take a shower." This suggests strongly that the cumulative effect of individual citizens’ conservation, and not government-imposed restrictions, is really making the difference.

I don’t think that the math behind either Bayes’s work generally (even if he never really applied anything) or the general idea of how to apply it here would be all that complicated. I would try to explain it myself, but I think that there are several good explanations out there already (google reveals a gaggle of them).

I also think that you are underestimating the use of the theory before the late 1990s. Remember that it is not just a tool for computers and caluculators, but the essence of a philosophy (arguments regarding causation – not its particular existence but the general nature of it – are bound, eventually or implicitly, to rely on Bayes).

Here, finally, is a good article for getting started along this line:

http://www.wikipedia.org/wiki/Bayes_theorem

The only reason, albeit completely unimportant in this context, for bringing any of this up, is that the careful understanding of the results of a frequency calculation and a Bayesian calculation are fundamentally different – not just more and less accurate.

Another article (whose implications for other cvillenews discussions, especially those involving law-breakers, are great): http://www.teacherlink.org/content/math/interactive/probability/lessonplans/bayes/home.html.

If you do bother (and you probably shouldn’t) to read that, look especially closely for reasons to doubt any part of the "Time Axis Fallacy" – a dubious theory that led me in particular to doubt almost everything taught in math departments about probability (but not about statistics).

true enough: restrictions are just a piece of paper or an announcement. restrictions don’t save water, people save water, through their actions.

let’s truly give credit where it’s due, though: local (and state) government drew attention to the problem, made it a media story, put out proposals and suggestions for water use, and spurred some of the water-saving activity through the restrictions. if local government had not taken up the issue, i really doubt much or any savings would have taken place because there would have been no widespread awareness of the issue. government does make important contributions. without them attending to the public good, we tend to be atomized little blobs looking to our own needs with little awareness of larger issues.

Well, since there is NO WAY IN HELL we are going to get 30 inches of rain, it’s pretty much a done deal that we’re pretty much screwed.We’re bound to get 30 inches

eventually. Our average annual rainfall is 44.34 inches, so in a mere…oh…9 months, we should be in good shape. You know, as long as we don’t use up any water between now and…um…July 2003.Lars, you make a good point — there’s a certain percentage of the population (you and me included, and I suspect a substantial portion of cvillenews.com regulars) that would like to be able to get all available data about this thing. I know that I’d feel better if I could look at some solid numbers. I don’t know if I could actually come up with any useful ideas, but I sure would feel better. Plus, if I could do the math myself to figure out exactly how screwed that we are, I may even find that I conserve even more.

"I don’t think that the math behind either Bayes’s work generally (even if he never really applied anything) or the general idea of how to apply it here would be all that complicated."

No, its not, thats why it was discovered years ago. The problem is the complexity increases quickly with an increase in the input. It’s simply a very "expensive" operation. Technology applies scientific knowledge in ways its discoverers never imagined. Newton never even imagined that his laws of motion would be used to put a man on the moon and orbit the earth! There are modern methods lumped on top here. I simply meant to say the way it was concieved and implemented was one way then, and a completely new way now. Because of computational speed.

Back in "the day" the manhattan project had to do a LOT of calculus, without computers or even pocket calculators. They had ROOMS full of people with pencil and paper, and they’d rush in with some results, and pass them down, and they’d all integrate at once, like some sort of demented orchestra.

So even though it could be understood and implemented, it was not popular. Without a computer it would be painfull. Explaining the calculations here would be simple enough, but it wouldnt produce results, this requires a significant amount of CPU time. So much CPU time, that the invention of the computer did not bring about any significant change the statistics business.

In the mid 1990’s large clusters and distributed network computing efforts began to provide enough power to solve the enormously computationally expensive iterative training required. Perhaps you’re thinking of the way they taught it back in the day. We’re not guessing weights here, we’re LEARNING them from relationships in the known data. This is exactly how the national weather service tracks front and storm tracks, through constant iterative machine learning. Machine learning was invented by another guy entierly, and thats a hotly debated question of who, so they just call it "ML" instead of "Bob learning" or whatever.

The machines are getting smarter, but it appears you’re still stuck back in college ;) Prepare to have yo’ ass assimilated to the borg, resistance is futile!!!

It is very powerfull for protein structure prediction, because simple homology (frequency) searches are likely to have too many matches with other sequences of unknown function in the databases. When sequencing can be done at your corner drug store in about an hour, the databases will grow faster than they can manufacture storage to put them on.

Dont worry, we’re on the mofo!

The "hard numbers" I’ve found are all at the state or federal level. There is a good amount of information here:

http://climate.virginia.edu/drought.htm

But I’d like some more fine grain local usage statistics. We won’t get it tho, that would be far too convenient :)

Charlottesville will be ok. If we need to pump water out of the river, we will definitly get that done. But that does nothing for wells. I’m lucky that Nelson isnt in emergency status like Charlottesville and Albemarle are. They made the "disaster list". Do you think that has something to do with the population growth in albemarle vs nelson?

The Commonwealth’s website says its a result of almost 10 years of below average rain. I find this hard to swallow, I was on a canoe in the james river in *9* feet of water just over a year ago. Now you can see rocks sticking out of the james every 2 feet. Other states have gone from dustbowl to "drowning cows and children in the street" in the same amount of time. The wool had been firmly pulled over our eyes. Its hard to see that way, and I’m confused, so I dont know why.

Sorry, I will shower everyday. Period.

Just take "military showers" – wet down, turn water off. Later, rinse. You’d be surprised on how little water you need to show if you do it this way. I’ll bet I probably use 2-3 gallons of water a day showering. Tops.

Here we go again….Now we find out that we’re up to 55.4% capacity in the resevoirs after the rainfall by 9/30…a level we haven’t seen since 9/22 or before. 1.9 inches of rain just bought us 8 days! Further, a couple weeks ago when we were averaging 10.6M gal/day, we were told we had 80 or more days left. Now that we’re down to 7.8M gal/day, and we’re back to 55.4%, how many days do we have left now?

Will somebody please please please report daily:

1) Percentage level at reservoir

2) Daily usage

3) How many days of water left

4) and put in the rainfalls as a side bar so we can see what that means too

And do this on the front page of some rag please! Or web site (Waldo…are you there?). The info is in all kinds of different locations…just collect it and put it all on one page or portion thereof so we can all measure this thing, cause as you know "You can’t manage what you don’t measure!"

Yes, this is a powerfull tool to keep usage down. If we are held accountable for our usage, we’ll all be scared into conserving more.

With Hurricane Lilli likely to follow the same track as Isidore, we may get enough rain to get us past Winter and into the Spring. Which means we will once again postpone the (in)decision.

Did you see all the runoff falling uselessly into drains? No way to store the water. Every large building in New York City has its own water tank. Every small house in Hamilton, Bermuda has a cistern to catch rainwater. What do we have?

Bureaucrats who think the solution to a water shortage is restrictions to reduce water use, instead of improving infrastructure.

To be fair, we only pay these people $9000/year.

You get what you pay for.

It’s Thursday October 3, and the level is 54.7% – and we have had no rain for a week.

This is why just looking at the level and extrapolating how mych rain we’ll need to get to 70% is really meaningless.

"Getting" to a specific level is not important. It’s NOT running out that’s important. Granted I’d be happier at 70%, but the fact is that water IS making it to the resevoir. And that’s a Good Thing.

I’ve seen no significant rainfall, and my lawn is lush and green, there are mushrooms (great way to tell if the ground is moist) that just came up this week. The rockfish has flow…. not much, but flow. Water is still seeping down through the system. It could easily take months for a water molecule in my yard to make it to the ocean.

The resevoir level is not based on rain, its based on consumption. Hasn’t anyone noticed that our population has doubled, and our reservoir is the SAME SIZE as it was before, being fed by the SAME creek?

Duh… its not the drought, its consumption. The drought just pointed this out to us. *SMACK* pay attention!