El Hierro eruption: climatic impacts?

On the ‘About’ thread new commenter Wyss Yim said:

An important key to understanding UK floods is what really drives the North Atlantic Oscillation. Natural or human-induced changes? For natural drivers we need to include submarine volcanic activity for switching on hot seawater. For example, the El Hierro submarine eruption off the Canary Archipelago from October 2011 to March 2012. See my article in Imperial Engineer Autumn 2013 issue.

Here’s the article written by him:

el hierro article

His hypothesis is:

Because sub-aerial volcanic eruptions and submarine volcanic eruptions are responsible for temperature, pressure and humidity changes, they may be important as triggers for weather-related events or patterns within the North Atlantic Basin.

In the article he then details the submarine eruption, lasting from October 2011 to March 2012, which apparently resulted in sea water heated to 18.8 degrees Celsius above normal from a total erupted volume estimated to be 329 million cubic metres.

He says:

The ‘sudden’ switching on of hot seawater in the southeastern part of the North Atlantic ocean has three main climatic impacts. First, the hot and low density seawater immediately beneath the surface speeded up the tropical Atlantic currents, rushing the tropical waters northwards. Second, the hot seawater warmed the atmosphere above causing a fall in air pressure to generate depressions. Third, the polar jet stream was drawn further south than normal in the North Atlantic Ocean. Consequently both the ‘normal’ oceanic circulation and  atmospheric circulation were drastically altered. The combination effect is an extremely negative NAO including the development of a ‘Greenland block’.

But hang on – where are the numbers to back up these claims? Or even references to papers explaining the data leading to these conclusions?  Being a curious sort, and having attempted calculations for volcanic heating before, I felt duty-bound to investigate, as they say.

Here’s an image showing the tongue of erupted material:

Digial elevation models presented by IEO showing the vent area before (a) the eruption and at present (b), with the new crater and even what is likely submarine lava flows. Images: Instituto Español de Oceanografía (IEO) via volcanodiscovery.com

Digial elevation models presented by IEO showing the vent area before (a) the eruption and at present (b), with the new crater and even what is likely submarine lava flows. Images: Instituto Español de Oceanografía (IEO) via volcanodiscovery.com

First a few calculations. From Prof Yim’s article (also found in other sources):

1. Heat released into the ocean by the eruption

Total erupted volume = 329 million m3
Density = 2.7 x 10^3 kg/m3 [1]
therefore mass (m) = 8.88 x 10^11 kg

Inital temperature assumed = 1200°C
Final temperature = 20°C (~ average sea temperature around Canary Islands)

Specific heat (Csp) of basalt = 1.50 x 10^3 J/kg/°C [2]
But since the lava will solidify, there is a phase change and we must also include the heat released – latent heat (Cla) of crystallisation = 5.00 x 10^5 J/kg [3]

Energy released Q = m x (Csp + Cla) x ΔT

For cooling, solidification and temperature reduction to 20°C of 8.88 x 10^11 kg lava the calculated total thermal energy released on cooling = 1.93 x 10^18 J

[1] Stephen R. Sparks J. & Huppert H.E. (1984) Density changes during the fractional crystallization of basaltic magmas: fluid dynamic implications. Contrib. Mineral. Petrol. 85:300-309
[2] http://www.engineeringtoolbox.com/specific-heat-solids-d_154.html
[3] Kojitani & Akaogi (1995) Measurement of heat of fusion of model basalt in the system Diopside‐forsterite‐anorthite, Geophys. Res. Lett., 22(17), 2329–2332, doi:10.1029/95GL02064.

From Nature:

Conductivity-Temperature-Depth measurements of the waters affected by the volcanic emissions revealed temperature and salinity anomalies of +3°C and −0.3, respectively, at 80 m depth and 290 m from the volcano. Maximum temperature anomalies of +18.8°C were observed over the crater at 210 m depth using expendable bathythermograph probes water acidification of up to 2.8 units within the first 100 m depth and 2 km from the volcano.

The eruption began on 10th October and ended in early March. Let’s call that 141 days, therefore we could have a heat energy release of  1.37 x 10^16 J/day.

The eruptive site started out at >200m and at the end of the eruption the cone was reported as 86m below the surface.  Let’s take an average of 100m depth, and imagine heating took place over an area of 1000m x 2000m, although that is much larger than the measurements reported.

2. Heating the ocean around the eruption

Volume of water = 2.00 x 10^8 m3
Mass water (m) = 2.00 x 10^11 kg
Specific heat of water (Csp) = 4,186 J/kg/°C

Calculated average water temperature increase ΔT = Q / m x Csp = 16.33°C

3. Transporting the heated water

What about the Canaries current – how would that affect the plume of heat?  You can see ‘the stain’ trailing away from the island on this Modis image:

The Canaries Current shown below is a cool current flowing southward.  From the Modis image, the visible part of ‘the stain’ estimated via the ‘ruler tool’ on Google Earth is about 35 km wide along the coast of El Hierro and leaves a trail ~2km wide by ~60km long on this image. 

The current has a calculated flow of up to 0.4 m/s. Let’s use a more conservative 0.1 m/s, equivalent to nearly 9km/day, and suppose the heat is carried in the top ten metres of water over a front of 2km.  That calculates out as moving 1.73 x 10^8 m3/day, which is very close to the 2.00 x 10^8 m3 that was the basis of the heating calculation.  That calculation suggests we have that volume of water at up to 16.33°C above ambient, not allowing for cooling moving in a 2km path southward.

If we use the figure for the current of 0.4 m/s. Then the estimated water volume moved is 6.91 x 10^8 m3/day, and as there is still the same input of heat, the temperature increase (ΔT) is estimated as 4.72°C. That’s much closer to the observation of +3°C increase.

The thing is – the article says:

First, the hot and low density seawater immediately beneath the surface speeded up the tropical Atlantic currents, rushing the tropical waters northwards.

Since the Canaries Current is a cold current moving south, does that mean to infer that the volcanic heat moving south and west in the Canaries Current eventually reaches the Western edge of the Gulf Stream? Hmm.

4. Effects on ocean heat, weather and climate

We have a ribbon of water about 2km wide and stretching for say >100km with a temperature of >16°C at the eruption site, and cooling to ambient along its length. The question is will that 200 km2 area have any discernible effect? I’m highly sceptical that it would have anything but a local effect. It is a drop in the ocean (quite literally).

The estimated size of the ‘hot plume’ (of +4.72°C) was 6.91 x 10^8 m3/day in comparison to the 3.46 x 10^11 m3/day of the Canaries Current – that’s 0.2% of the water in the Canaries Current, heated by 4.72°C….. or, we could have 1% of the water in the current heated by 1°C. Diddley squat.

From NCDC SST anomaly maps (1/4 degree resolution) – here are images for the five months of the eruption from 11 Oct 2011:

This slideshow requires JavaScript.

These have an insufficient resolution to pick up the local heat. October begins with warm water around the Canaries and African coast, which cools into Winter.  Warmer mid-Atlantic water  moves northeast through the Winter, and looking at previous and subsequent years, it is not possible to infer anything special about the period during the eruption.

The article says that the warm water in the Atlantic results in “an extremely negative NAO including the development of a ‘Greenland block’.”

However, looking at the NAO index for the period, it goes strongly positive, the negative later in 2012:

2011   10   0.94
2011   11   1.30
2011   12   2.25
2012    1   0.86
2012    2   0.03
2012    3   0.93
2012    4   0.37
2012    5  -0.79
2012    6  -2.25
2012    7  -1.29
2012    8  -1.39
2012    9  -0.43
2012   10  -1.73
2012   11  -0.74
2012   12   0.07

Moreover, the warm mid-Atlantic sea surface temperatures the article discusses are associated with a positive mode of the NAO:

Professor Yim blames everything from the busy Atlantic hurricane season to the record low Arctic Sea ice in Autumn 2012 on the additional warmth added to the Atlantic water from the El Hierro eruption.  Sorry, I’m just not buying it (unless of course my calculations have a huge error).

The article concludes:

In this look-back-and-learn analysis of the submarine eruption of the El Hierro volcano it is concluded that we have grossly underestimated the role played by submarine volcanic  eruptions as a trigger of weather-related events or patterns within the North Atlantic Basin. Unlike the conclusions drawn by a number of workers, it is unnecessary to attribute such weather-related events or patterns to anthropogenic global warming.

While I am happy to agree that such weather-related events or patterns should not be attributed to global warming, The only thing I learned was that this was an extrapolation too far.

This entry was posted in Volcanoes, Weather. Bookmark the permalink.

6 Responses to El Hierro eruption: climatic impacts?

  1. edcaryl says:

    Good work! I love it when you talk data!

  2. Bloke down the pub says:

    A bit OT but I wondered if you’d like to have your name on the side of a boat?

  3. Verity Jones says:

    LOL, thanks. A bit rushed, although the calculations used a spreadsheet from previous work so should be sound. I prefer analysis-based posts, but so little time these days.

    @Bloke down the pub,
    From the link “…asking members of the public to nominate 2,400 inspirational women – one for each mile of the race. For a donation of £42 their names will be printed on the side of Miss Hammond’s boat.
    If you’re inferring that I’m inspirational, I’m terribly flattered, but I’m sure there are many others much more worthy.

  4.  Doug Cotton  says:


    Doug, the rules under which you may comment have been explained to you. Since you chose to ignore them, do not expect any further comments from yourself or your aliases to be allowed.

  5. Tree says:

    Could you please clarify the maths you used here?
    I assume your formula for energy released (Q = m x (Csp + Cla) x ΔT) was just a typo, as it multiplies latent heat by temperature change. Allowing for this, your initial assumptions give 2.02 *10^18 J, resulting in an increase of over 17C at 0.1m/s. Where did the other degree go?

    • Verity Jones says:

      Thanks for spotting the error. I’ll check the calculations too. I used (adapted) a previous spreadsheet where the formula was used with a delta T of 1C, in which case it didn’t matter.

Comments are closed.