A new study by Steinke shows that the sun could have been a driver (somehow) of some of the monsoonal rain changes over the last 6,000 years over Indonesia and Northern Australia. h/t to The Hockey Schtick
In the spirit of the Perfect ClimateTM that existed prior to Henry Ford, we also find that Indonesia had a dry spell that lasted for a while, like say, 3,000 years. It ended about 800BC whereupon things got wetter, and mostly stayed wetter. The authors (Steinke et al) think this might have something to do with solar minima which was very low 2800 years ago. (Though I note the Greek Dark Ages also finished then, and “city states” arose, right, so it could have been that too. Ahem?)
To get straight to the action in Figure 6 the top squiggly line is AISM Rainfall (that’s the Australian-Indonesian summer monsoon). It shows how things were wetter in the last 2800 years ago and drier before that (annoyingly, the present time is on the left). The second part of the graph in red shows sunspot numbers. That gets flipped upside down and superimposed on the rainfall graph in the third part, and we can see a correlation that’s a lot like the CO2-and-temperature graph we see all the time, but is 5850 years longer.
Fig. 6. Changes in AISM rainfall and solar activity. (a) Ti/Ca ratios in core GeoB10065-7; (b) 10-year averaged reconstructed sunspot number (Solanki et al., 2004); (c) 10-year averaged ln-ratios of Ti/Ca (black) and sunspot numbers (red). 95% confidence intervals (in brackets) for the Pearson correlation coefficient (r) were calculated using a nonparametric bootstrap method, where autocorrelation has been taken into account (Mudelsee, 2003). 21-point running means shown in bold (aeb) to illustrate the long-term trends in ln-ratios of Ti/Ca and sunspot numbers. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.) (Click to Enlarge)
The big shift 2800 years ago is referred to as “drastic” and has been noticed in other places around the world:
Why the 2800 years BP event stands out so drastically in our monsoon record as well as in other records (see below), can currently not be answered conclusively.We suggest that in contrast to other solar minima, the 2800 yr BP minimum lasted longer than most other solar minima. This might have resulted in a stronger effect on the climate system, and also facilitates the detection in proxy records. Either way, there is clear evidence that the 2800 yr BP solar minimum affected climate conditions over much of the planet, including shifts of the Southern Westerlies (van Geel et al., 2000), the establishment of modern wind regimes in northern Africa (Kröpelin et al., 2008), and shifts in atmospheric circulation over Europe (Martin-Puertas et al., 2012). Moreover, the Dongge
Cave EASM record displays a decrease in rainfall around 2800 years BP but no change during other big events in solar output between 4000 years BP and 6000 years BP (Wang et al., 2005).
But nothing is simple. In far north west Australia stalagmites (Denniston et al., 2013, Fig. 5e) shows the “opposite behaviour of monsoonal rainfall with a decreasing AISM rainfall over the Mid-to-Late Holocene.”
They do try to explain the patterns, give them points for trying. It’s complicated. It looks like several big theories are going to be tossed out before this one is figured out:
In order to explain the generally higher rainfall levels after the 2800yr BP event we suggest that the combined effect of orbital and solar forcing is responsible for the long-term temporal behaviour of AISM rainfall over southern Indonesia as well as northern Australia. Despite an increase in austral summer insolation around 10 W/m2 between 6000 yr BP and 3000 yr BP, only a minor increase in AISM rainfall occurred between ~6000 yr BP and 2800 yr BP.We suggest that a long-term upward trend in solar output between 6000 yr BP and ~4000 yr BP (Fig. 6b) counteracts increasing orbital forcing such that the long-term trend in the Ti/Ca record is minor (Fig. 6a). After the 2800 yr BP event, enhanced orbital forcing keeps rainfall at a generally higher level than during the drier Mid Holocene. After ~1200 yr BP decreasing solar activity causes rainfall to increase
further for about 1000 years (Fig. 5a). The steady increase in rainfall after ~1200 yr BP is consistent with rainfall reconstructions based on dD of terrestrial plant waxes from Lake Lading (East Java; Konecky et al., 2013, Fig. 5b).
Overall, it’s interesting, but don’t draw too many conclusions. I thank the authors for their honesty. It’s fairly obvious that in the world of rainfall-propheses, scientists are struggling. This paper is filled (as it should be) with caveats and warnings as well as puzzling inconsistencies between proxy graphs of Australia, Indonesia, the Galapagos, and Ecuador. Things are just not neat.
Note the correlation is not too good, and the authors remind us this means a very vague amount — “some” of the variability is due to the sun.
In addition, we find a link between changes in AISM rainfall and solar activity with certain solar minima corresponding to stronger southern Indonesian rainfall, in particular at around 2800 years BP (see Fig. 6aeb). The correlation between the unsmoothed Ti/Ca and solar activity records is relatively low (r ¼ 0.319) but statistically significant for the past 6000 years (p < 0.05) when taking serial correlation into account (Mudelsee, 2003). The statistical significance of the correlation indicates that some of the variability in the AISM rainfall can be attributed to changes in solar activity (Fig. 6c).
I think what really limits conclusions here are the uncertainties in timing of things that happened so long ago, the Egyptians were building pyramids. It’s just too long ago. I mean, I presume if those dates are out by 20 years, the correlation gets shot to pieces. Imagine in the year 5050, trying to match up 1994 rainfall to solar activity in 2014?
Fig. 4. (a) Titanium (Ti), (b) iron (Fe) and, (c) calcium (Ca) intensity XRF logs in counts per second (cps) of core GeoB10065-7. (d) ln-ratio of Ti/Ca of core GeoB10065-7.
Here’s a tiny bit of technical info to give you an idea of how they figure out rainfall so long ago. Figure 4 shows the minerals they analyzed. They claim that increased amounts of Ti and Fe mean there was probably “an increased supply of siliciclastic material” from the rivers. The ln-ratio of Ti/Ca means lower rainfall to wash sediments in to this spot that was being analyzed. The Iron (Fe) is there as a check of some sort.
What about ENSO?
The authors try hard to look at El Nino and La Nina proxies as well, but end up concluding that thing are very not clear and rather than being causal, something is possibly driving both rainfall and the ENSO pattern. Frankly given that El Ninos etc have such a big effect in the space of a few months, it seems admirably ambitious to try to correlate things so long ago. Dare I say “Brave”?
It is expected that more frequent and/or intense El Niño events have resulted in reduced rainfall and subsequent
drought in the AISM region and, consequently, less riverine terrestrial supply to our site. Comparison of our Ti/Ca record with the lake sedimentary records from Ecuador (Moy et al., 2002) and the Galapagos Islands (Conroy et al., 2008) shows no correlation [r ¼ 0.089 with 95% confidence interval (0.296; 0.134)] and a high covariance [r ¼ 0.481 with 95% confidence interval (0.237; 0.605)], respectively. However, despite the
high covariance, comparison of our Ti/Ca record with the El Junco Crater Lake in the Galapagos Islands (Conroy et al., 2008) reveals that periods of more frequent and/or intense El Niño events after w3000 yr BP (Conroy et al., 2008) are associated with increased terrigenous supply and thus enhanced AISM rainfall (Fig. 8).
However, since El Niño events cause reduced rainfall and subsequent drought in the AISM region, the positive correlation between El Niño events and southern Indonesian rainfall, in particular after w3000 yr BP, does not imply a causal relationship, but perhaps a common forcing.
How does this solar connection work?
The authors do their best, and use their own models to try to figure out a hypothesis that explains all the puzzles. I’m grateful that they use the word “hypothesis” and don’t make any glorious claims. Though I don’t feel obliged to take those models seriously for a moment ; we all know how dismal climate models are on rainfall.
I wish them the best of luck. You get some sense for how tough their job is here. El Nino’s ought to make things drier over Indonesia, but according to one big theory (I see the name “Mann” in there below) solar minima ought give us El Nino’s, yet solar minima seem to inspire buckets of rain in Indonesia instead:
The most conspicuous shift in terrigenous sediment supply and thus AISM rainfall occurred at around 2800 yr BP, coinciding with one of the strongest grand solar minima of the Holocene (see also above; Solanki et al., 2004; Usoskin et al., 2007). As lower solar radiative forcing is usually associated with less surface ocean evaporation and, consequently, reduced monsoonal rainfall in tropical regions (Meehl et al., 2003), our finding of enhanced rainfall over southern Indonesia during times of reduced solar activity, in particular the 2800 yr BP grand solar minimum, seems counterintuitive. Moreover, according to the theoretical mechanism of a Pacific Ocean “dynamical thermostat” (Clement et al., 1996; Mann et al., 2005; Marchitto et al., 2010), solar minima should favour El Niño-like conditions and hence drier climate over Indonesia. To find a possible mechanism that could reconcile reduced solar activity with enhanced southern Indonesian summer rainfall, we analyzed the output from an idealized solar sensitivity experiment (Varma et al., 2011) using the coupled climate model CCSM3 (Collins et al., 2006). In this experiment, solar forcing is simply implemented through a change in total solar irradiance (TSI) with no wavelength-dependence and mostly affects the climate system through shortwave absorption by the surface. The TSI has been reduced by 2 Wm2 (corresponding to 0.15%) for a period of 70 years to capture the multi-decadal timescale of typical solar grand minima (Usoskin et al., 2007).
So file this away as another paper reporting a tantalizing connection, and note that there are some signs of hope that some modelers are adding in long term solar factors and looking for a connection.
a b s t r a c t
The Australian-Indonesian monsoon has a governing influence on the agricultural practices and livelihood in the highly populated islands of Indonesia. However, little is known about the factors that have influenced past monsoon activity in southern Indonesia. Here, we present a ~6000 years high-resolution record of Australian-Indonesian summer monsoon (AISM) rainfall variations based on bulk sediment element analysis in a sediment archive retrieved offshore northwest Sumba Island (Indonesia). The record suggests lower riverine detrital supply and hence weaker AISM rainfall between 6000 yr BP and ~3000 yr BP compared to the Late Holocene. We find a distinct shift in terrigenous sediment supply at
around 2800 yr BP indicating a reorganization of the AISM from a drier Mid Holocene to a wetter Late Holocene in southern Indonesia. The abrupt increase in rainfall at around 2800 yr BP coincides with a grand solar minimum. An increase in southern Indonesian rainfall in response to a solar minimum is consistent with climate model simulations that provide a possible explanation of the underlying mechanism responsible for the monsoonal shift. We conclude that variations in solar activity play a significant role in monsoonal rainfall variability at multi-decadal and longer timescales. The combined effect of orbital and solar forcing explains important details in the temporal evolution of AISM rainfall
during the last 6000 years. By contrast, we find neither evidence for volcanic forcing of AISM variability nor for a control by long-term variations in the El Niño-Southern Oscillation (ENSO). 2014 Elsevier Ltd.
REFERENCE
Stephan Steinke,*, Mahyar Mohtadi, Matthias Prange, Vidya Varma, Daniela Pittauerova, Helmut W. Fischer (2014) Mid- to Late-Holocene AustralianeIndonesian summer monsoon variability, Quaternary Science Reviews 93 (2014) 142e154