THIS NOTE HAS BEEN REVISED AND UPDATED IN JULY 2021
UPDATED 9 MARCH 2018
Recently there have been a lot of media reports about the present water shortages in Capetown, South Africa with the latest “Day Zero” being 9 July 2018. Photographs of near-empty reservoirs and queues of residents lining up to collect water from public taps suggest that lack of rainfall may be the cause of the problem. Or could increasing water demands, leakage losses, poor water management and inadequate future planning also contributing factors? Shortages in water supply can be due to many causes, but surely population growth and increased water demand must also be significant factors? The population of Capetown is now about 10x that in 1930 and 4x that in 1970 (these two dates being when severe droughts affected the Western Cape). Of course, most of the media casts blame for the current water supply problems on drought caused by global climate change!
I have analysed the annual rainfalls at stations located in Capetown. The only data that I have been able to locate are those on the KNMI (Royal Netherlands Meteorological Institute) website. However, I have recently found some published plots of Capetown rainfalls which were used when no other data were available.
Three records of monthly rainfall were located in the Capetown urban area: Royal Observatory (1850-2000), Pinelands (1900-1995) and Capetown D F Malan (the International Airport) (1951-2017). Unfortunately the data since 1998 are very poor with many months of incomplete or missing data; with seven years since 1998 with no annual totals. Additional data for Royal Observatory (now SA Astronomical Observatory) for seven years since 2000 were extracted from plots in a publication by Dr P Wolski of UCT (University of Capetown). Attempts to obtain original rainfall data for the period since 2000 were unsuccessful.
Here is a plot of unadjusted annual rainfalls (November-October) for the three stations:
Using simple correlations, the data recorded at Pinelands and at the International Airport were adjusted to equivalent annual rainfalls at Royal Observatory:
It is seen that the annual rainfall trend has remained essentially constant with a succession of droughts and wet periods, each often over periods of three to five years. The current drought in Capetown commenced in 2015 but similarly severe droughts have occurred in the past, in particular the droughts of 1926-1931 (5 years) and 1971-1973 (3 years). Other severe droughts in Capetown occurred earlier in 1851-1854 (4 years), 1864-1866 (3 years), 1894-1897 (4 years) and 1963-67 (5 years).
A more detailed comparison of the annual and monthly rainfalls during the droughts of 1864-1866, 1927-1931, 1971-1973 and 2015-2017 indicates that the characteristics of all three droughts were similar. These four droughts were chosen because the annual rainfalls were less than 500mm for three consecutive years, 500mm being about 20% below the long-term mean annual rainfall. The concern is that the 2015-2017 drought has not yet ended and therefore, based on the 1927-1931 experience, its continuation for another year or two would not be at all unprecedented.
Please let me know what you think of my analyses. brigun@westnet.com.au
The Author:
This article was written by Brian Gunter of Narooma, NSW, Australia. In his previous life Brian was an engineering hydrologist involved over many years in the analysis of rainfall and river flow data for the planning of water resources projects in Australia, Asia and Africa. In recent years he has been one of the Marine Rescue NSW (previously Royal Volunteer Coastal Patrol) volunteer weather observers who operate the Narooma station for the Bureau of Meteorology.
Published by
Brian Gunter
Retired civil engineer.
Particularly interested in climate trends.
View all posts by Brian Gunter
Climate Trends 
Nice observations. I am an amateur who has spent a good deal of time over the last 10 years studyinh climate related material. One of the important folders is my rainfall folder. I had noticed the story about the impending zero water hour back in January. So I looked up material to read, and that is how I came across your site.
I lean heavily on what I see as cyclical processes within the climate system to formulate assessments. Recently, I have done well, imo, of interpreting correlations between solar/rainfall correlations. So I was curious to obtain a decent lomg term graph such as what you present above.
At first glance, and then a longer look, my impression was that there was no correlation between the solar cycle and the Capetown rainfall graph. This was interesting in itself, and I chalked it up to differences in the SH versus the NH. I came for another look as I have thought about your graph since then.
Once again it was difficult getting a foothold, but then I focused on the beginning of the graph which was easier to visualize. Next I paid attention to the drought years which you highlight in your post. All of those droughts correlate with the downslope of the solar cycle after the maximum has been reached. The longest droughts shown occur during what I term as a cool trend, in this case the years from 1885/86 to 1916/17. Then 30 years of warming take hold up to 1946/47. That is then followed by 30+ years fo cooling up to 1976/77. The warming commenced once again up to 2005/07, and a cooling trend commneced since then. We are still in the cool trend which will now make itself apparent in the years ahead.
Now with that in mind, look at when your longest droughts occur, “…1894-1897 (4 years) and 1963-67 (5 years). …”. Both of those droughts occur on the downside of a solar cycle, and during a cool trend. We are now close to solar minimum, and we are in the middle of the cool trend just as in the above cases.
I am responding mainly to add weight to your question “… The concern is that the 2015-2017 drought has not yet ended…”. I think that you are correct for the reason which I have stated in this comment. I have an unusual knack for reading graphs, and I have been able to use that ability to correctly forecast a number of weather/climate related events over the last 4 years, in particular, as I made what I consider a breakthrough in understanding some parts of the climate system back in early 2014.
ps…I also find your graph even more interesting than before. Although I can now tease out the solar cycle component, that only highlights that there are other factors affecting weather patterns in the SH. In the NH, I can readily read rain/snow graphs from many disparate areas of the NH, and find the solar cycle correlation. I am going to give your rainfall graph a deeper look to see if I can come to a more indepth understanding of weather/climate processes in the SH, and thus a better picture of the globe as a whole.
LikeLike
My graphs are simply presentations of recorded raw rainfall data, where regular cycles of drought and wet periods are seen to occur at most locations. Mostly these extend over periods of 3-7 years so are not random annual events. There must be some cause for these “abnormal” events, be it solar or something more complex. I have not tried to come up with an explanation of these periods, but merely point out that they do occur. Your correlations with solar periods are therefore of particular interest.
Your comments were particularly related to Capetown rainfall where we have a continuous record since 1850. I have also presented similar annual rainfall plots for other stations in South Africa as well as for stations in Australia and the US – see my other blog articles. I have also made similar plots of mean annual temperature for many stations worldwide and note the strong cyclical trends at many locations.
While I dislike the common approach of trying to estimate the “world average” rainfall or temperature, I am also aware that there can be wide variations in single geographical regions. For example, in South Africa there are considerable variations in the rainfall trends at locations such as Capetown, Port Elizabeth and Kimberley. Note that at Cape Leeuwin in the extreme SW corner of Australia there has been an ongoing drought since 2001. I would think that variations in ocean currents, global atmospheric patterns, distance from the coast, elevation, topographical features, etc may also affect rainfall trends. In other words, weather and climate are complex systems, but that does not mean that solar influence may not be a major factor.
LikeLiked by 1 person
My outlook is that the primary cause of the patterns which we record are modulated by the oceans. Obviously the Sun feeds the oceans, but the Sun also has cyclical patterns such as the Gleissberg cycle or the occasional grend solar minimum which lead to warm/cool shifts here.
Then surface winds are also of importance as they push water and warm/cool air masses to different areas. I have been payinh much attention to surface winds of late. The North Atlantic surface wind changes are very interesting. I think that earthnullschool is of good enough quality that it can be used to witness the current climate shift in action, when observed daily over time. It looks to me as if the SH is in a cooling mode as surface winds around Antarctica have shifted to a new pattern at the end of Janyaray which is still dominant today. That pattern is pushing cold surface waters and cold air masses northward up the west side of SH land masses. Imo, if this pattern holds for the next several years to come, then you are looking at future cooling. …https://earth.nullschool.net/#current/wind/surface/level/overlay=temp/orthographic=82.40,-54.84,672/loc=50.927,-79.800
LikeLike
Also, another takeaway from my observations is that the SH always leads the planetary climate shift vs the NH, and the lead is somewhere around 6 to 10 years.
LikeLike
I see that you are from Australia. What do you think of this analysis of the Onlsow Tropical Cyclone record? …https://wattsupwiththat.com/2018/02/27/study-extremely-stormy-weather-in-california-happened-over-150-years-ago/comment-page-1/#comment-2753898
Several of the regulars at WUWT plainly stated that my thoughts are nothing but wishful thinking, including Dr Svalgaard whose solar cycle chart I used in making my analysis. The reason for their severe doubt is that I am an amateur with no maths skill. My ability is that my reading comprehension is still very good, and I have a superior ability to interpret graphic representations derived from science studies. I stand by that claim as I have been able to correctly forecast/predict future pieces of the weather/climate puzzle. That makes me an advanced wiggle matcher, imo.
Believe it ir not all of the above is a consequence of my having grown up in California to a father who encouraged me to learn how to fish. That led me to leave San Francisco, and move north into the mountains where the steelhead and salmon were. That led me to learn in the early 1970s that there was a suspected cyclical flood pattern on the West Coast of the US. Being a Californian who had lived through two of those massive rain winters, 1955/56 and 1964/65, that idea intrigued me.
That sole bit of possible knowledge became my foothold into the climate debate story, when I started reading about AGW in mid 2008. That bit of knowledge became a spark when in my first year of reading I came across my first look at a solar cycle graph. Within seconds my inner thoughts said “Look at how the flood history of the West Coast matches up with solar minimums”, and just like that I was hooked. That is how I work. My mind immediately saw the correlation even though my only science experience was 1 year at university in 1969 for chem/physics of which I never completed and then d ropped out. Now I fully realiize that I always belonged in the fields of science. I will spend the remainder of my years thinking science.
Lastly, after all of that my greatest prediction was in early 2014 and naturally had to do with the West Coast hypothetical flood pattern. I had a period of great clarity. That moment crystallized the knowledge from studying AGW over the years, and I gained a level of understanding of what I had been looking at for the prior 6 years. I made comments shortly after that the winter of 2016/17 was the most probable year for a flood level winter on the Pacific Northwest, and I nailed it, 3 years before it happened. Around the same time, I also stated that the winter of 2017/18 would be the downturn into the next cool trend, the point where the trend on a graph moves to the shouder of a downward step at which time it becomes readily apparent. That means that I also had the ENSO region conditions, and the solar cycle position correct in order to correctly make the flood prediction. And that is why I think that I am on the right track despite what some professionals may say.
LikeLike
Here are some basic numbers as to the correlation between sunspots and Capetown rain patterns.
Drought yr Sunspot Cycle Solar Max Solar Minimum
1851/54-4yr #9 1850 1856
1864/66-3yr #10 1862/63 1867
1894/97-4yr #13 1892/93 1900/01
1926/31-5yr #16 1926/27 1932/33
1963/67-5yr #19 1958/59 1965/66
1971/73-3yr #20 1968/69 1975/76
2015/?? #24 2013/14 my guess 2018/19
I should have posted this at the beginning, but I get a little excited sometimes. Last thoughts, one more year of drought for certain, highly probable for 2 more years as this look like the mid 1960s as an analog, imo.
LikeLike
It is interesting to see the correlation between solar cycles and rainfall. However, droughts are not general worldwide at the same time (eg droughts in Western Cape but perhaps floods in Japan) so there are obviously other major factors that affect our climate. But it seems logical, and there is some definite evidence, that solar activity and astronomical factors affect the observed historical cyclical trends in rainfall and temperature.
LikeLiked by 1 person
These correlations which I see are indeed limited to local specific areas. Drought in the Capetown area. Floods in the Pacific Northwest/California. Possible tropical cyclone activity at Onslow Australia.
I don’t find the correlations everywhere. What got me started on all of this was that I knew that there was a hypothesized cyclical flood pattern for Northern California/ Pacific Northwest rivers. Then when I saw a graph depicting solar cycle history my mind immediately saw a correlation. That was the hook back in early 2009 which I couldn’t resist. That led to a multi faceted successful prediction made in January 2014 for the winters of 2014/15, 2015/16, and the heavy rain/flood winter of 2016/17. That also included a correct prediction of the state of the ENSO regions, and the position of the current solar cycle #23 being close to minimum in 2016/17. Now I have to wait until 2026/27 to see, if I really have something. That should be the next iteration.
So that really sparked my thoughts, and off to the races I went. At the time when I first started talking about this back around late 2012, and saying that I thought that the next solar minimum would occur around 2018/19, the consensus was that the solar minimum would occur around 2021/23. Dr Svalgaard, who correctly predicted that SC 24 would be a low cycle, said no way for 2018/19 back then. I say that as he was one of the few solar physicists in the world to correctly see that SC 24 was going to be a weak solar cycle. The consensus at that time from all the rest of the best solar minds was that SC 24 was going to be a big cycle, and the world would roast due to global warming. So that was a bit of an achievement for me, a total amateur, to connect the dots successfully. The only person in the world to do so. All due to my thoughts on the California/PNW possible flood cycle. Think of what this will mean if in 2026/27 there is another West Coast flood. It would give the solar folks a new tool for projecting all future solar cycles. It would give those who study the ENSO regions a new tool for predicting La Ninas, and that is very important as that affects rainfall/crops worldwide.
Then there are the potential other cyclical patterns of importance for other areas of the world such as South Africa or cyclone activity around Western Australia. All of that is my reason for sharing what I see.
Note that I live in a trailer park in the mountains of California. I live on Social Security and food stamps. In other words I do not seek financial gain for all of this, even though that would be a possibility. My gain is in sharing something which could help many others. Keep your fingers crossed that all of this holds up. And thanks for sharing your thoughts which also sparked my imagination.
LikeLiked by 1 person
We’ve been told that the current drought is a “once in 300 years” event!
Have you come across Will Alexander’s (not the musician!) work?
Sample: https://journals.co.za/content/civileng/49/2/EJC26999
LikeLike
Thanks Vuurklip for the reference to Alexander etc. I will check it out.
Whether it is a 100 year or 300 year drought is always dependent on what data are used. Also it depends on what period you are thinking of, eg 1 year, 3 years or 5 years. In my analysis I used the unique long rainfall record at Royal (now SA Astronomical) Observatory where records exist since 1850. This kind of data does not exist, as far as I know, anywhere else in Africa! Also I only looked at the 3 year rainfall average as 2015-2017 is only 3 years. In 2020 I will be able to do the same thing for the 5 year average!
My 3 year analysis shows that the 2015-2017 drought was the most severe since 1850, but only slightly so. The 3 year average rainfalls were 399 mm (2015-2017, 419 mm (1971-1973) and 455 mm (1927-1929). So the current drought is severe and, despite a bit of recent rain, is far from over.
LikeLiked by 2 people