Accurate and Precise

Scientists are always concerned with the quality of the data they collect. The precision and accuracy of any measurements must be good enough to resolve whatever phenomena they are designed to measure. There is always a balance between the cost and time needed to get more accurate and precise data versus what is needed. Usually, scientists try to match the accuracy and precision with the task at hand. No measurement will be perfect, but the goal is to ensure that any errors are minimal and are much smaller than the signals we are trying to measure with our instruments.

In science, the word “accurate” means "capable of providing a correct reading or measurement." A measurement is accurate if it correctly reflects the size or value of the thing being measured. On the other hand, “precise” in physical science means "repeatable, reliable, getting the same measurement each time."

The Focus of our Research

The main focus of this research expedition is to get a better idea of how the shape of Axial Seamount changes in response to magma entering and exiting the system. These inputs (magma being fed from deep below the volcano) and outputs (eruptions or magma intrusions) change the shape of the volcano. When magma fills the magma chamber the volcano expands, inflating like a balloon; but during an eruption the volcano surface lowers and deflates. Ideally, our measurements of these changes are both accurate and precise. That would ensure that the data we are collecting can be used to constrain models of what is happening in the subsurface.

At Axial Seamount, we measure the precise depth of the seafloor (the volcano’s upper surface) by measuring the pressure on the bottom caused by the height of the ocean above the instrument. On this expedition we are measuring pressure in several different ways.

The following instruments get us a pressure measurement at a single point with an error on the order of +/- 1 cm.

Continuous measurements:

Reccovering the BPR instrument.

1. Mini Bottom Pressure Recorders (Mini BPR) – placed on benchmarks by ROV Jason and left for a year or two. They record continuously and keep a digital record for the entire time they are in the water. There are 5 of these. 
2. Moored Bottom Pressure Recorders- are dropped from the ship and record for a year or two similarly to the Mini BPRs. We recover these by sending an acoustic release code, and the BPR floats back to the surface to be collected by the ship. We have recovered and then redeployed 4 of these on this expedition. 
3. In addition, there are 4 Cabled Bottom Pressure Recorders (Cabled BPR) attached to the Ocean Observatory Initiative (OOI) cabled observatory at Axial Seamount. These send their data through a fiber-optic cable back to shore, which can be seen in real time at this website: www.pmel.noaa.gov/eoi/rsn/

One point in time measurements:

Two pressure recorders on AX-308 benchmark.

1. Mobile Bottom Pressure Recorder (MBPR) (carried by Jason to various points (benchmarks) on the volcano). The purpose of this measurement is to be able to correct for drift in the continuously recording devices, and thus be able to measure slow gradual depth changes over several years like volcano inflation. 
2. AUV re-mapping, described in our last blog. Allows us to extend the measurements of inflation at Axial Seamount over a larger area outside the caldera (where the BPR data is limited to). These long swaths are less accurate and precise than the BPR measurements, but the error (+/- 20 cm) is still small enough compared to the size of the depth changes to give us useful data on the inflation or deflation of the volcano.

AUV bathymetry is shown in color.

The combination of all this data collected in several different ways allows Scott Nooner and Bill Chadwick to model the location, depth, and shape of the magma reservoir inside the volcano. Their modeling of the magma reservoir is consistent with the locations of earthquakes at Axial.

The Effect of Drift

A challenge of working with the continuously recording BPR’s is that their readings tend to drift gradually with time. That means they are still precise, but their accuracy degrades with time. During this expedition, one of the main things we are doing is taking the MBPR around to all the benchmarks on the volcano with Jason making repeated measurements during a long survey extending over several days. These measurements help us constrain the amount of drift in all of the BPR instruments so that we can measure the gradual inflation of the volcano between eruptions. Because the pattern of inflation and deflation is similar from one eruption to the next, these measurements also help us forecast when Axial is ready to erupt again (see: www.pmel.noaa.gov/eoi/axial_blog.html)