Today over at Volcanocafe, the author Albert wrote a great post about volcanic inflation. It’s a truly great read, and this post is meant partially as a response to his article.
In particular, I wanted to discuss inflation styles, and how inflation caused by dike intrusions is different from deep inflation.
Can Volcanic Inflation Be Used To Predict Eruptions?
Volcanologists are constantly trying to predict future eruptions so that a minimal amount of destruction occurs. While inflation is a strong indicator of future eruptive behavior, it’s not always as straight forward as it may seem. For example, Grimsvotn had very little inflation prior to its eruption in 2011. Other volcanoes such as Yellowstone vary greatly in their inflation and deflation rates with nothing exciting occurring. And on the other hand, there are situations like Sinabung, where inflation predated future eruptive behavior.
The main problem is that volcanoes don’t always show lots of inflation prior to a big eruption. There have been three mid-sized eruptions in Chile since 2008 (Puyehue Cordon-Caulle, Calbuco, and Chaiten), and none of these exhibited strong inflationary trends prior to eruption. Going further back in history, the largest eruption of the last 50 years at Pinatubo showed very little inflation prior to the main eruption.
With respect to these volcanoes that don’t seem to have major inflationary trends, there WAS some inflation. But the problem is that the inflation was very sudden, very dramatic, and occurred only a short time prior to the onset of eruptive activity. So what makes this type of inflation unique?
Shallow Inflation from a Dike
The biggest takeaway from these events is that the rapid inflation came from the dike intrusion that lead to the surface. Since dikes are inherently narrow, and since the intrusions came close to the surface (later erupting), the region of inflation is not particularly wide. Also, due to the speed of the advancing intrusion, inflation can be quite dramatic in a short period of time.
Given, not all dike intrusions erupt. It may be more common for these types of intrusions to stall in the crust and not erupt than it is for them to reach the surface to form an eruption.
So does a narrow region of inflation + rapid onset mean there is a dike intrusion occurring?
While dike intrusions are a strong possibility, it’s important to keep in mind that these deformation patterns are also possible due to changes in the water table and hydrothermal fluids below the surface. With that said, it’s definitely a strong possibility that a finger of magma is pushing its way up from the primary magma chamber when you see inflationary patterns that reflect a rapid change in a narrow area. See the graphic interpretation of this below.
Broad Inflation From Below
On the other hand, you have quite a few examples of volcanoes that are slowly inflating across a very broad region at a somewhat steady rate. Broad inflation is more simple in that magma being added to the magma chamber slowly causes inflation over a long period of time. Since the magma chamber is much larger than a dike and much deeper, the inflation tends to push affect a wide area of crust. Since the inflation comes from the expansion of the magma chamber itself, the change is very slow and gradual even on a geological time scale.
This broad inflation can potentially trigger future dike intrusions or changes in the hydrothermal system above it (as a result of increased heat), both of which can result in eruptions.
One of the main problems for broad type of inflation is that there isn’t any great way to predict how far a magma chamber can inflate before it needs to relieve pressure via eruption. See the graphic interpretation of deep inflation below.
Why do some volcanoes seem to not inflate at all prior to eruption?
Some volcanoes seem fairly resistant to showing any inflationary trends. Grimsvotn doesn’t tend to show any major change in inflation, but that doesn’t mean there is no change. There are still noticeable signals in the GPS that signify changes in the magma system below, but these changes are not always vertical – they can often push the land above to the side just as much as the inflation would push the land up above it.
On the other hand, a volcano like Pinatubo which didn’t exhibit noticeable deep inflation was likely already primed and ready to erupt as a result of past inflation and expansion of the primary magma chamber that was not measured. Remember, deep inflation often occurs very slowly over a long period of time, so just because a volcano isn’t currently showing signs of deep inflation doesn’t mean there is a lack of magma ready to erupt below.
One Final Caveat: Assimilation
Something that is often overlooked when discussing eruptible magma and inflation is the assimilation factor. In most volcanoes, the magma comes from deep within the mantle, often being the product of decompression melt, a subducting slab, or a hotspot. But this doesn’t account for the fact that additional magma may be generated when the surrounding crust is melted and joined with the current magma pool.
This is especially relevant for very large and very hot magma systems that are not sitting on oceanic crust. The Taupo volcanic system, the most prolific volcanic region in the world is notorious for producing a lot of its explosive rhyolitic magma by melting the crust with hot basalt magma.
The point to consider here is that when magma is assimilated, it may not be accounted for by inflationary changes since there won’t be significant net volume changes since this doesn’t involve adding new magma, but rather just changing the bedrock into a melted and eruptible state.
Some Sources and Further Reading: