Benjamin Franklin as far as I recall.

That's why it makes very little sense to blow trillions of pounds on something which is far from certain and looks more and more unlikely.

The same with Sao Paulo, money would be best spent increasing storage and reducing pollution rather than thinking the current drought is the new norm and building desalination plants. Ask the Australians about desalination plants and permanent drought..

That is quite a bet.

1) Increase storage and you gain nothing if rainfall remains low, for you have no way to fill the extra storage.

2) Build desalination plants and you waste your money if the rains return.

If you get 1) wrong you are left with an uninhabitable city.

If you get 2) wrong you are out of pocket. You still have backup for the next drought.

I would choose the desalination plants.

stewgreen
There'll be

EM
By the looks of things, more storage and less pollution looks a better today.

Sao Paulo could do with some rain. They recorded their hottest day on record on Friday.

stewgreen
Mr, not really that much of a pedant though, Mrs S pointed it out when reading over my shoulder.

Entropic man
On the other hand the USA seems pretty clear of drought, or would we be panicking about that too if records began in 1934?
USA Palmer Drought Index
compared with 1934
USA Palmer Drought Index 1934

Those areas of California are a known result of ENSO which climate models can't predict or model.

Sao Paulo hottest day, these things have a habit of swings and roundabouting

Location

This report describes the historical weather record at the Campo de Marte Airport (São Paulo, Brazil) during the last 12 months. This station has records back to January 1973.

São Paulo has a warm temperate climate with dry winters and hot summers. The area within 40 km of this station is covered by grasslands (43%), built-up areas (23%), croplands (13%), shrublands (12%), and forests (6%)

The longest cold spell was from October 28 to November 9, constituting 13 consecutive days with cooler than average low temperatures. The month of August had the largest fraction of cooler than average days with 61% days with lower than average low temperatures.

This goes back to 1973, I couldn't find a station going back as far back as your link claims, 1943, although what little is available to me doesn't give any detail on continuous records, UHI due to urbanisation and all the other problems with city temperature records.

EM, what is a digital thermocouple. All thermocouples are analogue. It might be that the voltage measuring circuit uses A-D conversion but this doesn't make the thermocouple digital!

On a more interesting topic you say that you regard most proxies as valid within their limitations. In this respect I like to do a thought experiment such as I have a wet bulb thermometer that has no radiation shield. I don't have any independent data for humidity, not for direct solar insolation on the thermometer. Neither do I know when the bulb is wet, or if it has inadvertently dried out at times because the water reservoir wasn't filled, or evaporated to dryness. I have several years, maybe many years, temperature readings from the thermometer. How do I go about reconstructing the air temperature?

This problem is rather like that of using tree rings as a proxy for temperature. I'm not sure that it can be done with any precision what-so-ever.

Any suggestions?

Jim Bouldin made some very interesting posts on the problems of obtaining temperature data from tree rings - I think it eventually got up to 14 posts.

http://ecologicallyoriented.wordpress.com/2012/11/10/severe-analytical-problems-in-dendroclimatology-part-1/

In the very first post he makes these points which seem to me to render invalid any temperature reconstructions using tree rings:

"(1) ring width, being the result of a biological growth process, almost certainly responds in a unimodal way to temperature (i.e. gradually rising, then rather abruptly falling), and therefore predicting temperature from ring width cannot, by mathematical definition, give a unique solution,
(2) the methods used to account for, and remove (“detrend”) that part of the long term trend in ring widths due to changes in tree age/size are ad-hoc curve fitting procedures that cannot reliably discriminate such trends from actual climatic trends, and
(3) the methods and metrics used in many studies to calibrate and validate the relationship between temperature and ring response during the instrumental record period, are also frequently faulty."

I cannot see why these reservations should not also apply to using trees as rain gauges.

Coincidence ?

Entropic man
More likely part of the same natural cycle, careless in a dry country 2001 or 1869-72 and an abrupt shift to drier conditions at ca. AD 1400 is contemporaneous with a change from sedentary village life to regional desertion and nomadization (sheep/camel pastoralists) during the preindustrial era in formerly Ottoman realms, and thereby adds climate change to the multiple causes for Ottoman Empire “decline.”

You don't look hard enough for precedents?

There is a presidential election on October 26th. Probably more heat than light in the local media until then.

Bloomberg now reports Cantareira at 3.3% and due to run out in mid-November. I hope stewgreen was right about the rains.

SandyS

I think the question is not whether such droughts have occurred independantly before, but whether there is a new factor linking them

I was browsing droughts in 2013 and 2014.

California, Spain, Italy, Syria, Jordan, Iran, India, China in the Northern Hemisphere.

Brazil, Kenya, Sahel, Australia, New Zealand in the South.

Except for New Zealand, they are all just beyond the usual latitude boundary between the Hadley cells and the Ferrel cells. Could the common factor be an expansion of the Hadley cells?

Entropic man
Is it a new thing though?

SandyS

Now there's a $64,000 question.

A quick google search "expansion of hadley cells" found this . The abstract describes "a robust poleward expansion in all seasons".

There were also a large number of pdfs which I did not pursue, with titles suggesting a similar pattern.

Since on land the boundary tends to be marked by desert, changes in the desert boundaries in the past would be informative. You could then look for correlations with past temperatures.

Paul Dennis

You wet bulb thermometer challenge was fun.

First the boundary conditions. The highest temperatures would be dry bulb thermometer with high air temperature and insolation. That sets an upper boundary for air temperature, though insolation would make it higher than the actual maximum air temperature.

Below 0C the water will be frozen, so effectively you have dry bulb temperatures. The lowest temperatures will probably be at night, with no insolation, so you could get a quite reliable minimum air temperature.

Together these give you the annual temperature range.

To derive average wet and dry bulb temperatures, try a frequency distribution. The graph should be bimodal. The lower peak would be the wet bulb average and the higher peak the dry bulb average.

Accuracy? Probably +/- 10% of the range.

EM some interesting observations that you could test by downloading several years of wet bulb thermometer readings from any met station. Of course the highest temperatures are only dry bulb temperatures if there is no longer any water left to wet the bulb. That might not be the case. Below zero the wet bulb thermometer may actually be insulated and so not produce true minimum temperatures.

What is +/- 10%. Do you mean you can determine the mean annual air temperature to within +/- 10% of the range recorded by the thermometer? I think that's optimistic but let's say the range is 40 degrees C. That would be typical for my garden on an average year. So the mean is determined to +/- 4 degrees C.

Such an estimate is useless in the context of palaeoclimate reconstructions and I suggest that measuring tree ring widths is no better than, in deed probably a lot worse than, just having wet bulb thermometer readings.

I enjoyed your thoughts but I contend that your answer only shows that it is not possible to record past temperatures from such under constrained systems with any degree of accuracy or precision. It is not possible to record past temperatures from such under constrained systems with any degree of accuracy or precision. I chose the wet bulb thermometer analogy because there are similar factors involved with regard to tree rings - temperature, insolation, water availability and stress etc.

Entropic man
It's a bit of a problem then, as reliable records pre-20th century for deserts and the interior of Africa and South America are a bit thin on the ground. Exploration of Africa by Europeans only got off the ground mid 19th century, Victoria Falls 1855 and no one had traveled the entire length of the Niger River until 1946.

Until proven otherwise I'll go with natural variation which is not unprecedented.

Entropic man,
I don't know about you but I think this has been worth moving from Unthreaded?

On reflection perhaps records of the Doldrums/Horse Latitudes might be a more fruitful method of checking how things have changed over time. There must be a reasonable number of ships logs available for research, but even they wouldn't go back far enough to make a reasonable guess in my opinion.

Not all Brazil is suffering apparently, still a disaster no doubt.

That’s the message on Tuesday from Brazil’s Ministry of Agriculture, Livestock and Food Supply, marking the start of the 2013-2014 harvest season. The ministry anticipates a record year for Brazilian agriculture and the agribusiness sector.

Brazil is bullish on its agriculture.

The ministry forecasts a record harvest of 90 million tons of soybeans, which could help it overtake the U.S. as the world’s top soybean producer. The 193 million tons of projected harvested grain also moves Brazil closer to the ranks of the world’s top food producers, a circle dominated by the U.S., China and India, among others.

www.usda.gov/oce/commodity/wasde/latest.pdf
From WUWT

SandyS

It has been a good discussion.

I hope those agriculture results are accurate, and not hyped to sounds good.

This is the day before a presidential election I doubt that anything said by a Brazilian politician at this point can be considered reliable!

Let us give it a week or so to filter out the campaigning effect and see what they say then.

As my wife comes from Southern Brazil and I have visited the area a number of times, this discussion thread caught my eye. Before linking the drought to climate change you need to consider the following geographical facts.

- The City of Sao Paulo is built on a plateau about 700 metres above sea level. This means that although the Tropic of Capricorn passes through the North of the City, the highest recorded temperature is just 35.3 °C.
- The two principle reservoirs have been supplying the city since the 1920s when the population was less than a million, compared to twenty million today.
- The area of Sao Paulo and Rio de Janeiro was once covered by the Atlantic forest. This once extended well south into Argentina and inland to Paraguay – an area bigger than Western Europe. According to Wikipedia 88% has disappeared. From extensive travel in Parana, Southern Sao Paulo state and Northern Santa Catarina this figure seems accurate.
- The Rivers of Sao Paulo State mostly form part of the River Plate Basin, that meets the sea in Montevideo, Uruguay.
- Sao Paulo State is about the same area of the United Kingdom and has a population of 43 million.
- Upstate there is intensive agriculture, including soybean, sugar cane and cattle.
- This is the worst drought in 84 years, not ever recorded. That is four years before the American dust bowel of 1934.

My conclusion is that the seriousness of the current water crisis is due to the following factors, in order of importance.
1. Investment in water supply not keeping pace with demand.
2. A once-in-a-century drought.
3. Location of a megacity on a plateau, limiting the ability to cheaply extend the water supply.
4. Changes in rainfall patterns from deforestation.

Kevin Marshall
As someone without a great deal of knowledge of Sao Paulo, and many other mega-cities round the world, 1,2 and a variant of 4 would be my initial posit for any city about to run out of water. I would guess Number1 applies to the South East of the UK, and should there be a repeat of 1976 then London would be facing similar problems to Sao Paulo. It is forty years since Denis Howell worked his magic so a once in 50 year drought in the UK could be expected in the medium term?