Rainfall

At the start of the season we we warned that it was going to be a very wet one. Lot of rain and heavy falls. I know it happened in Natal but here on the Highvelt it is dry. The blue is the long term average, red last season in green this one so far. One thing I learned from my farming years is that you have to take the forecast with a very generous dose of salt.


Alas, I don’t have MS Office on my computer, and when I try to open the document with OpenOffice, I just get a blank page with nothing on it.

But I am rather skeptical of long-range forecasts too. Perhaps climate change plays a role? Purely intuitively, it seems to me we are long overdue for a drought here in Pretoria; we used to get one every few years. I can’t even remember when last we have had a summer drought though. We seem to be among the global warming winners.

Ditto on the blank page, but I have to say in joburg it seems we’ve not gone a day without some kind of thunder-shower in the last month or so.

How do I change a Ms Office graph into a picture? This rainfall is measured at my house in Middelburg and if you can see it you will understand why I feel sorry for the farmers. On the other hand maybe you are right that a dry year is overdue. The farmers can mostly all afford Land Cruisers anyway, so maybe this won’t harm them too much.The weather service still got it wrong though.

Accompanied by some spectacular electrical storms. Its normal Jo’burg weather though, bleeding hot until mid/later afternoon when it thunders and rains.

[ol]- Select the Excel graph by clicking on its border.

  • Click the “Copy” option on Excel’s “Home” tab.
  • Open MS Paint (Start —> Programs —> Accessories —> Paint).
  • Click the “Paste” option on MS Paint’s “Home” tab.
  • Save the picture (I recommend in PNG format).[/ol]

'Luthon64

Thanks M. Let’s try.


http://www.telegraph.co.uk/topics/weather/9010030/South-Africa-weather-forecasters-threatened-with-jail-if-predictions-wrong.html

Well, well!

Typical of the somewhat confused way in which the media report things. The heading implies that any weather forecaster who gets it wrong will be imprisoned, which is simply not true, but of course, the heading is what most people will read and brainlessly repeat.

In the local press I have seen it breathlessly reported that if your husband is on his way home and you phone him to warn him that there’s a big storm brewing and he should be careful, you can be sent to prison for it. Predictably, a, er, big storm ensued over it and the weather bureau had to hastily backtrack and issue statements and who knows what else, all over media reports that anyone should have been able to see were probably at best half true.

It is probably perfectly sensible to regulate weather reporting, in order to prevent needless panic or doomsday prophets freely predicting the world will end next week when a superstorm hits Jo’burg. Perhaps the law is even written in such a way as to imply you’re not allowed to phone your husband to tell him there’s a big hailstorm breaking over the neighbourhood, but I think it is pretty obvious that no one will ever be prosecuted for that.

Maybe a possible way around this is already being employed when predicting rainfall. Rather than saying that it will rain or that it won’t, the weather service estimates the chance of precipitation. So, if the chance that a devastating storm with flooding will occur is, likewise, reported as a percentage, I don’t see how any weather service can be held accountable when the day in question turns out bone dry. :confused:

Rigil

Nah, don’t like this at all, simply because it is yet another onslaught on my rights to say what I please. Rather, people should be taught how to discover for themselves the difference between good and suspect sources of information.

Rigil

However, that probability is quite precisely defined and, in point of fact, fairly reliable. When the Weather Bureau cites a “probability P of rain,” then that figure is relevant to a certain area or region A for a certain time period T, and means that it will rain over fraction P of area A within time period T. These are the limits of accuracy/reliability of our current meteorological models, which are known to be of a class called “chaotic dynamical systems.” These models are extremely sensitive to tiny variations in their input parameters, and typically produce vastly different results for very slightly different inputs.

It therefore has little to do with hedging bets on the part of the Weather Bureau.

'Luthon64

It does :o?

I thought that it meant that there is a P chance that rain will occur at any given point in area A during T.

Please see http://www.srh.noaa.gov/ffc/?n=pop

(Which I took to imply that it is at least possible that no precipitation could occur anywhere in the area).

Rigil

If you think about it, the two are saying the same thing. That’s because within A and T, there is no preferred point and time at which it will rain. That is, any point and time within A and T is statistically equivalent to any other, and so the probability of having rain at any randomly chosen point within A and T is P.

'Luthon64

No, I don’t immediately see the equivalence of the two interpretations,

The first claims that it will rain over some fraction of the area, the second that it needn’t.

… but i’ll go away and and try and work it out first before becoming too confident. :confused:

Rigil

Take a simple shape like a square or a circle. If the chances of rain are, say, 30% then 30% of that area are shaded blue and the remainder red. Pick another simple shape to represent the blue area. Pin your drawing to a dartboard. Put on a blindfold and throw a dart at your drawing. Assuming your dart hits anywhere within the total area, what is the probability that you hit blue?

There is, of course, a small residual possibility that there is no rain anywhere within A during T, but if the Weather Bureau cites a chance of 20% or more, I wouldn’t put any money on it not raining anywhere over the relevant area and period. That is the nature of probabilities.

'Luthon64

I’ve checked this in a simulation - attached, hope it opens - and it turns out that if the probability of rain at each point in the forcast area is equal to P, then the percentage of points that experienced rain after the simulation is also jolly close to P! 8) Unfortunately the math as to why it should be so is beyond me. :-X

Rigil


This is short term forecast. My original stance was that this was supposed to a a El Nina year but as you can see on my graph this (here in M’burg at least) is definitely not the case.

I became very skeptical about the SA Weather when they forecast a 80% chance of heavy falls and there was not even a cloud in the sky on the day. I was still farming then and we needed the rain.

Good job Rigil (you star), but does it convince you that the two views are equivalent? More importantly, has it refined your understanding of what those rain probabilities mean?

The maths, as in most real-world problems, go back to calculus: Consider an area A and shrink it progressively so that it gets closer and closer to zero without ever quite getting there. The assumption is that a sub-area is representative of the whole. (Meanwhile, forget that the process is an idealisation!)

Tweefo, long-term weather prediction is, with today’s technology, a pipe-dream. Meteorological models are continuously being refined with advances in mathematics, computation and empirical understanding. It’s not an easy problem.

'Luthon64

Yes and yes. :slight_smile:

Rigil