….. should have been about.
Who would have thought back in January when we were deploying a set of seabed platforms onto the seabed to the west of the Azores, we’d be writing a blog about the challenges of home working?
How times, and our lives, can so radically change in such a short period of time.
As the Easter holiday approaches we should have been doing the preparations to go and recover these instruments, but instead we are having to leave them there until …….. who knows when.
Fortunately the part of our platforms that enable us to recover them from the seabed – the acoustic release – has a battery pack that will power it for 18 months.
So as far as the Azores instruments are concerned we don’t have to worry about them for quite a while yet.
The scientists involved are also quite happy as they are going to get a data set far longer in duration than was originally planned which, when recording earthquakes, is always a good thing.
Longer-term blog readers will also recall that we deployed a set of instruments in a deep sea canyon offshore the Congo in September last year. These instrument platforms are due for recovery in October and so they are not an immediate problem.
The current COVID-19 situation has presented a number of challenges though, even to a group that are used to having to be entirely flexible to any situation, who regularly work remotely, and who are used to having to respond rapidly to changes in circumstances.
Even though all seagoing operations have been cancelled, we like to look at the positive aspects of everything – even COVID-19.
Our blog readers will know that not only do we deploy instruments of a diversity of kinds onto the seabed to underpin a broad swathe of science applications, we also design and develop a lot of that instrumentation, bespoke to the science project itself.
We have a long list of developments that we’d like to do on our “To do” list that we normally get little chance to do because we have to prioritise the things with immovable deadlines, like deployments at sea, or things we have to build so that they can be deployed to an immovable seagoing deadline.
So having some time “at base” gives us an opportunity to do all of the developments on our “To do” list – or at least make some significant progress on them.
What complicates this further – is that “at base” now means at home with the various team members scattered about, but still being resourceful to enable us to carry on.
At the moment we are currently working on two things:
a) the “To do” list
b) a repair/rebuild of a set of equipment that belongs to the National Marine Equipment Pool in the UK.
This latter set of equipment is not ours, but being a skilled and versatile lot, we can turn our hand to it as it shares many of the same concepts of operation that our seabed instruments do.
So having been asked to fix it ready for a test that is planned for later in the summer, we have taken on the challenge and risen to it.
And while we are fixing it, we are going to improve it to:
i) make it work much better
ii) be better constructed so that it is far less prone to damage
iii) build in a means to do data and systems quality control on-the-fly as data are being recorded and the systems being used at sea.
The latter involves making junction boxes to connect things together, making sensor networks to make measurements (e.g. air pressure) in a variety of places within the various systems, making data “harvesters” that progressively extract data from systems as its being acquired and then stores it in a central data storage area; plus making a set of web-based tools that interrogate those data files and check not only that what’s in them is correct, but also checks the system performance against metrics or thresholds, or against some global datum or standard.
Divide and conquer is the motto here.
So the team are all doing various parts to build the collective solution.
One of the team is making the junction boxes – in this case these are also intellegent ….
…. having first constructed a work bench at home.
Everything we make like this follows a standard form:
grey plastic box with seal – keep water and moisture out;
a variety of connectors – all of which seem to be very expensive!;
a PCB brain;
something to connect it to a computer network (the yellow thing in the image below is an internet interface); and
a lot of wires.
Nice tidy job this one – and this one will harvest measurements of pressure from a network of sensors mounted at various points in the air generation and delivery system required for a towed seismic source.
Separately, somewhere else working at home the Boss has created the corresponding QC tool to take these pressure measurements and display them over time as a day elapses …..
…. and even to do the stats on these measurements so that any variation that might indicate a system problem can be spotted, with the idea being that a fault in a component can be fixed when it develops, rather than when the entire system fails.
In the stats image below the target system pressure is 2000 psi (pounds per square inch) and the fake dataset being used for development purposes shows how such a display might show where the average pressure is too high or too low, or where the variation in pressure is far broader than it should be.
These sorts of display can reveal air hose leaks, a problem with a compressor and so on.
This air generation system feeds a seismic sound source which is driven by a separate system that records times, depths and what the sound source looks like.
We are having to repair that too – and this one has a lot of cables running in/out of the back.
The sound waves made, that enters the seabed and bounces off the sub-seabed geological layers, are ultimately recorded by a series of sensors towed behind a ship in a long snake-like tube called a streamer.
This is run by quite a sophisticated system with quite a few parts to it.
We’re going to have to refresh its “brain” and give it a new perspective on life.
The Boss, with the aide of one of her colleagues, has already built the QC tool for the data this system records, which scientists look at in two ways:
- to create a picture of what the geology of the sub-seabed looks like,
- to look at the characteristics of the returning sound waves, as from this they can tell what the system is doing as it is towed through the near surface waters.
Here is a sub-seabed view from the QC tool, using data from an earlier research project we were involved in.
It is what the crust of the Earth looks like in the middle of the Atlantic at 13N where the American and African plates are moving apart at about the speed at which your finger nails grow.
As the crust is just being “born” there, there is no sediment, but some very amazing fault structures.
The other way to look at this data is to look at its frequency content – what are the frequencies of the signals in it – a frequency spectrum
The dips – or notches – in this one, at the blue lines annotated with the number 8 – tell scientists at what depth the sound array was towing – the 8 here meaning 8m.
There should be a second notch for the streamer, if the streamer were towing absolutely level along its entire length – the one in use here was 3000m long.
The depth of a streamer is controlled by a system called a bird controller.
What this does is to change the angle of a series of vanes attached to the streamer that either cause it to tow deeper or shallower through the water to drive it to the prescribed flat tow depth along its entire length.
The bird controller we were asked to repair turned out to be beyond repair not just by us, but even by the manufacturer and is having to be replaced.
But that doesn’t mean that we can’t build QC tools for it, and a harvesting box to extract that data.
Based on the format prescribed by the team member making the harvester, the Boss has come up with this tool, which plots the depth of the streamer at each vane – or bird as they are known; plots the water temperature and battery voltage of each bird and creates a map of each bird’s heading relative to magnetic north.
The latter let’s us see what the geometry – or shape – of the streamer is as it is towing being a vessel – side currents can cause it to have quite a bend which needs to be corrected for as part of data anlaysis.
When viewed on a computer, the above display is dynamic and changes on each cycle of information received from the harvester.
The above is fake data created just to test the tool, whereas in real operation the goal is for all birds to tow at the same depth; the battery level to be constant and not diminishing quickly towards the threshold, and the control vanes to have a zero inclination.
In the latter case, this means they will not be having to work to maintain streamer depth which causes them to strum and, in turn, creates water-born noise around the sensors in the streamer itself; and of course consumes battery power!
So despite being forced by circumstances into working at home – we are keeping quite busy and productive.
And when not working on work-related things we are also keeping ourselves amused by turning our skills to other things, like creating a “turbo” for a normal road bike just to get some exercise. So from:
a) two rotten fence posts
b) off-cuts from monitor base boards made for the at-sea tests for the above
c) kiddy buggy wheels
d) some bearings
e) a couple of pieces of Dexion shelving
f) an off-cut of steel bar
….. one turbo …..
…… which has been tested up to 40 km/hr.
Happy Easter blog readers.