How to select the right CTD [Webinar]

How to select the right CTD


Picking the right CTD for your application is critical. But what makes each CTD excel in different deployment scenarios?

In this webinar, we walk you through common technical requirements and considerations for specific applications and what CTD options are best for each one. From sensor integrations and sampling strategies to environmental factors and cost, discover what questions can’t be missed in your decision-making process to get the most out of your CTD.

The video below is a recording of the live presentation and includes the interactive Q&A with participants. Hear questions from users about their specific application and benefit from our team’s practical advice. Read the transcript, browse related content, or contact us for more information.

Short on time? Skip ahead to learn:

  • the first thing you should think about when selecting a CTD (2:00)
  • why not all CTDs are ideal for the same application (7:25)
  • important physical considerations when selecting a CTD (16:25)
  • extra considerations when selecting a CTD  (18:43)
  • more about RBR CTDs

Questions and answers

What’s the first thing you should think about when selecting a CTD?

A: How you plan on deploying it, since this will determine the specifications. For example, for profiling work you’ll want a CTD that can sample fast. (2:00)

Can I use one CTD for all applications?

A: Not usually. While CTDs are versatile instruments, they are often tailored to a specific application. Size, shape, and power are common considerations, particularly for glider integrations. (7:25)

What are the most important physical considerations when selecting a CTD?

A: It varies case-by-case, but some common ones include: maximum deployment depth; if you’d like to measure other parameters (such as oxygen and chlorophyll a); and if you need realtime data. (16:25)

What additional considerations need to be included in your decision-making process?

A: We recommend asking the following questions:

  1. Specifications – does the CTD’s data quality match what you need to answer your scientific questions?
  2. Maintenance – how often does the CTD need to be calibrated? Can you do the maintenance yourself? How long will the CTD be at the factory for maintenance (if needed)? How much will maintenance cost? (18:43)
How can you maintain your RBR CTDs?

A: RBR CTD maintenance is designed to be simple! Before every deployment, check 1) the desiccant, 2) the batteries and 3) the o-rings. We recommend annual calibrations, where we also check that the CTD is still operating as it should. With a three-week calibration turnaround, you can quickly get back into the field. (43:36)

How can you get the most out of your CTD?

A: We posted a series of CTD tips on YouTube, including using Ruskin‘s Autotask feature to set up multiple instruments with the same configuration and using Wi-Fi to easily check your data when in the field – without opening your CTD!

Where do I find the RBR docs site?

A: The RBR documents site can be found by clicking here.


View the transcript

Hi, I’m Candace Smith. I’m the product marketing manager here at RBR, and today we’re going to talk about how to select the right CTD. We’re going to talk about some technical considerations, so particularly about the features of a CTD that you should think about; some other physical considerations, some of which are still technical, but you’ll see what I mean by those; and then extra considerations, which I didn’t really know what to call them – it’s not the technical and environmental, it’s more bigger picture things.

I’m going to show you a whole bunch of pictures very soon about how to deploy a CTD, over in the chat feature. If you want to throw in the coolest way you’ve ever seen a CTD deployed, I think I would find that very interesting. That would be really awesome. I’m going to show you a whole series of images – about 12 different types of deployments. So again, if you want to share a cool or odd deployment, that would be really fun.

The motivation behind this was, I was a salesperson for quite a while, and being in sales, a big part of my job was actually coming up with and recommending to users how to pick a CTD. So, a lot of this stems from that, from the types of questions that I would ask somebody about their deployment or about their environment that they were deploying in, to be able to give them a recommendation for an instrument that will work and get the data that they need for their project to be successful.


So that’s really where the motivation from this stems from. Okay. The big question, I think the most important question here, is how are you planning to deploy the CTD? Typically, when people come to us, they have an idea about a research program or something they’d like to try, or a project in mind. So, it’s usually an easy question to answer. And now I’m just going to show you a bunch of examples.

The first thing you should think about when deciding to pick a CTD is: how are you planning on deploying it? These folks are deploying it through ice in a profiling application. So that’s one way. This is a picture from NOAA, where a profiling float is being chucked out of the back of a plane before a hurricane to do profiles. That is a very different way to deploy a CTD, because there’s a CTD on this float. This is an observation platform here at COVE. That’s where I’m based, here in Halifax. So, this is an observation platform. It stays at one location on the bottom. Other people call these landers. And in this particular case, this is a cabled one, so it’s a real time observation system.


So again, a little bit different than the other ways that we’ve seen deployment so far. This is a pretty common one. I would call this a small boat deployment. You can see they’re very close to the water. They’re just putting a CTD over the side and into the water.

This is something to think about: are you doing it from small boats and zodiacs? Are you putting CTDs on Argo floats? Is that how you want to deploy your CTD? In this particular case, these folks are deploying through ice like the first image, but also with real time data being sent up to the surface because it’s being frozen in. This is, again, an interesting one. Or are you deploying it from a cage? Again, there’s different ways.

This is a glider, there’s also a CTD on this glider. So again, that’s a different way of deploying than, say, using it through ice. Are you going to be putting it on a platform like a profiling platform? This image is of a Wirewalker. It’s a platform that moves up and down the water column and takes a profile on the way up. Now, we can see a CTD down here, on a rosette from a ship. That’s another way you can deploy CTDs. That’s a pretty common one, just like from a small boat.


This is our research scientist, Mat Dever, during his time at WHOI, deploying a CTD from a winch. So that’s another way of deploying a CTD. And finally, a very common way is deploying CTDs on a mooring line. This is somebody setting up a mooring line. You can see a few CTDs on here.

So again, if there are any deployment types that I missed or that you’ve seen done or that are of interest to you, throw them into the chat. So really, that was many, many different ways of deploying a CTD. But here at RBR and I think in most places, we really would define those as three different applications.

And I’ve said the word a few times: profiling. So just for everyone’s benefit, I know for most this is probably preaching to the choir, but profiling really means you’re going out some way onto the water and you’re doing a profile. You’re getting from, say, the surface down to a particular depth and back up, CTD information to get temperature, salinity, density, depth…

These are all kind of similar and we think of it as the same sort of application, they’re all profiling. We saw the example at the very start, through ice. It’s like a day trip: you’re going out doing a profile, moving along a small boat or zodiac. Maybe you’re doing a whole series of stops. When you put it in a cage, typically you’re deploying it as profiling.


Often cages are used on bigger vessels so that as it’s going down, it’s not smacking on the sides. A Wirewalker, because it’s doing this clean profile on the way up, that is a profiling application. And from a rosette, that’s a very common big ship deployment – the rosette is going down, you’re getting a clean profile on the way down. And then on a winch, again, a nice clean profile on the way down. It could be, you know, a lot of these are in conjunction with something else. If it’s on a rosette, you’re really doing it on a large ship.

Moored CTD observations are actually a little bit different. The sensor technology can be the same, but it’s a different kind of measurement in my mind, in the sense that you’re putting the CTD at one particular location and getting a time series of data instead of a profile through depth.

So here, this is where we have the observation platform, this CTD that I showed you earlier from COVE. Again, through ice. The folks who were doing the long-term moored deployment through the ice, those instruments are never going to come back. So, they have to send the data in real time. And one of the most simple or the most common ways of deploying a CTD is on a mooring line. It’s typically many CTDs on one mooring line or on a buoy.


And then we have integrations. For integrations, I’ll go into it a little bit more later. But essentially, we saw three in those quick series, like deploying a float from a plane. So, there’s on the float, on an Argo float (which is an international program of profiling floats), or on a glider. Those are integrating CTDs into platforms. Whereas with profiling, the most examples that I showed you, the CTD itself is sort of the platform to collect all the data. Okay.

At RBR, our CTDs are designed with a purpose. The purpose is if its profiling, if it’s moored, or if it’s an integration. So, if you have a profiling CTD, putting it on a mooring line might not be the best. If you have a CTD that was specifically meant for being moored, it might not be super easy to add it to a float.

I’m putting this out there as a PSA: really use the CTD for the original intended purpose. If you are calling us, “I’m going to be in the Antarctic and I want to deploy through an ice hole, etc.,” and add all these features and functions to it, and then you’re trying to deploy it in a different type of situation, call up your sales rep or support and just ask, you know, is it possible to use it in a different way? Because they are very much designed with purpose in mind.


That’s sort of the beginning.

When you’re getting or looking for a profiling CTD, these are things that you should think about, from my opinion. Again, you might have different ideas. Oh, and I haven’t asked you, sorry, we’re also going to do a poll. I should have said that at the beginning. There’s going to be a poll about how you deploy. There’re two questions in the poll. Keith will launch it.

I’m going to go through what I think are the most important things to think about when you’re buying a CTD or getting a CTD. And there are other things that I might miss. So, it would be really interesting to hear your thoughts about what you consider to be important when picking a CTD.

So, a profiling CTD. The very first thing I like to think about is: capturing the structure. And by that, I mean the dynamics or if there’s stratification. I’m a physical oceanographer by training, so I think about that piece: can I see the thing I’m trying to see? Can I do the science? For a profile, you really want to be able to capture the structure. So again, probably preaching to the choir here, this is a super simple example of a temperature profile, a salinity profile, and a density profile.


And here you can see the thermocline, the halocline, and the pycnocline. And that’s the reason that profiling needs fast sampling, capturing many measurements every second – sub-second sampling, and fast responding sensors, sensors that are going to respond very quickly, – because of these features like the thermocline, the halocline, and the pycnocline, because you want to sample fast to capture this fast change. If you’re only sampling once a minute, you might miss this whole feature. You might get a point here, and then here you’re not really seeing where the thermocline is, so you really want to be able to capture that structure.

We use Wi-Fi, but having a way to check that you actually got the feature that you were interested in is a little bit reassuring for me. Say, you’re going out for the whole day, you’re going to do this thing, and is there a quick way just for a sanity check to make sure, did I get the data? Did I capture this front that I was trying to find? Did I see the thermocline? Did I hit the bottom? You know, often we’re doing these profiles to answer a science question. So, if you’re trying to get all the way to the bottom or you’re trying to go through the pycnocline with a fluorometer to see if there’s a chlorophyll maximum there, all those things.


It’s just really nice to be able to check some way quickly that you got the data right. And finally, one thing I think about a lot is, you know, will the CTD last me for the day? Will I have to change batteries out in a Zodiac, like the Zodiac we saw earlier, filled with people? Is that going to be doable or easy?

If you are doing fast sampling, like 8Hz, is there enough memory on the CTD without having to pull out my computer and download it? Is there enough battery to last me for the day? Or maybe I’m going out on a cruise and I’m going to be out there for a couple of weeks. Will the instrument actually collect enough data? So really, if you think about collecting data from a 16Hz versus a 2Hz perspective, if you have a 1m/s fall rate on the CTD, that’s the difference in one metre of water of getting 16 data points, or two. So again, when we think about these physical features in the water column like the thermocline, the halocline, if you’re sampling faster, you’re going to have a better chance of capturing those features. So having 16 data points versus two, you’re much more likely to see that data and have that high resolution data.


That’s just a particular example of how that looks. Okay.

So, for moored CTDs, again, I have three things I like to think about. The first is related to the sampling rate. You know, analogous to the profiling one, it’s the deployment length; it needs to last for the entire deployment. And with moored CTDs I usually talk to people who are looking for somewhere between a month up to two years, that’s typically what I’ve heard. So, you really need the deployment to last the entire time of your deployment, right? You need the CTD to last.

You can also think about how you want it to be collecting data. Again, with profiling data, usually you want to sample as quickly as possible to get as much data as possible. For moored CTDs, I find you really want to maximize the sampling. So, if you’re out there for two years, you’ll sample so that you know the batteries or the memory will last up to two years. If you have enough battery for another three months, you might want to tweak it to get the most out of the deployment. And you can also think about things like bursting and averaging. Often, we’ll set instruments up for what we call continuous sampling: you take a sample once a second or every minute, every month, whatever it is…you can’t do every month, but you know what I mean.


You can also do bursting or averaging: you can do a burst instead, of the instrument turning on for one minute; you could say turn on for five seconds and do that every minute. So, you’re getting more data. It might help with any spikes in the data that you’re seeing.

The second thing is the number of units. Typically, on a mooring line, you’re deploying more than one instrument. If you have multiple instruments, how easy is it to set them all up exactly the same? Is that doable? Is it doable in a short amount of time?

And then power is always going to be a big question on buoys. Is there enough power for the CTD or other instruments for the whole length of the deployment?

And finally for the integrations. The integration piece is actually at most companies called “OEM” (original equipment manufacturer). That really means that we work directly with the manufacturers of the platform, with the glider manufacturers or the float manufacturers, and we give them essentially just the sensor piece, and we work really closely with them. When you’re picking a CTD for a platform, you first want to make sure that the platform is right for you. If you’re trying to find a glider, there might be autonomy requirements – like how long can the mission go, depending on what you’re trying to sample, – or depth requirements.


We actually do integrate on a bunch of different platforms. Power is probably the most important piece from an integration piece. These are fully autonomous platforms, so they’re not right now – I’m not aware, sorry – right now, I’m not aware of any of these platforms being able to get power from another source while they’re out there. Having low power on the sensors is really beneficial, because it will extend your deployments. The lower in power the sensors, the longer the deployments. And again, these are fully autonomous systems. They’re out there by themselves with no way of getting more power. So typically, the lower the power, the better.

And then the size. These aren’t huge pieces. You can see the scale here from Rutgers  – this is a glider, and this is a human. Sensors can’t be gigantic to fit on here.

Okay. So those are the considerations, kind of the technical considerations for the different applications that I think of, for profiling, moored, and integrations. The next part is these other physical considerations.

I think the second question after “how are you planning on deploying your CTD,” “how are you trying to use it,” is “how deep?” Because the deeper options typically require metal, typically titanium, because plastic materials just can’t withstand that kind of pressure. That’s something to think about. And make sure that you’re deploying to what CTDs are rated to: if it’s rated 200m, send it to 200m. Really make sure you’re deploying them to the correct depth that they’re rated to.


Another – I call this a physical consideration, I guess – is what other non-CTD parameters are you interested in? I’ve been talking solely about CTD so far today, but there are many other different types of sensors that people want to integrate now, and we’re getting more and more requests for biological, chemical, and geological sensors. Thinking of the bottom corner here (a CTD with extra options to measure oxygen and turbidity or something else), having just the CTD is often great and awesome, but it’s not always exactly what you need for your science question.

Think about if you need real time or near real time data transfer. And then with that, you have to think about things like cables and connectors and how easy it is to integrate those things. There’re also ways to transfer data using different kinds of modems.

Then there’s environmental factors. This includes questions like, are you expecting biofouling? Are the currents really strong? If it’s on a mooring line, do you have to worry about, e.g. maybe I should get a smaller CTD because it’s lighter, there’s less drag, and to make sure that it’s really secure to the line. And temperature is going to be a big thing for anything that’s battery powered.


We saw a couple of examples of polar work earlier. Often in those situations, especially for, say, the mooring line, you do need more batteries because temperature, cold temperatures, really affect battery life. You need to consider that and consider: if you are deploying it in ice or freezing conditions, will the sensors still work?

I call these the extra considerations. I didn’t really know what else to call them. They’re just other things to think about. One is how good is the CTD? And that might seem silly, but I think it’s a really important thing. If you think about the specifications like the accuracy, the resolution, and the stability of the sensors, that’s really important. And I’m going to get into that a little bit more. You could tick a lot of the boxes like the fast sampling and that might be fine. Or for example, you have a CTD that will last for your whole five-year mooring campaign. But, depending on the stability or the drift of the sensors, what are those data going to look like in five years?

A calibration turnaround is a big one we hear a lot about: what is the cost of not having your instruments in the water? How much of a cost to you and your science is not having it in, if you have to pull your instruments out and wait a few months to get your instrument back?

And the ease of use: do you need to be very, very qualified and especially trained in ocean technology to be able to operate these systems? Or can you teach somebody how to use them in a quick couple of hours?


And the consumables: the consumables are typically things like the batteries and the O-rings. The things that wear out over time. You want to think about: are they common to get or are they proprietary? Do you have to go back to the manufacturer to get them? And if not, still to think about, can the user replace them, or does it have to go back to the factory to get done there? That’s something to think about.

These aren’t physical things or feature related. This is more about the actual manufacturer itself and how it deals with calibration and used pieces. And these, I find, folks don’t think about as much. People do ask, what’s the maintenance schedule, so they can put that into their budget. But there’s other pieces to think about that are a cost. There is a cost to this as well. I put that in quotes because there’s lots of costs out there.


The three specifications I like to think about are: (1) accuracy. That’s essentially how close to the truth is the measurement. It’s probably my favorite, my favorite parameter. I don’t know what that means. But if you have a pressure sensor that’s saying, okay, I’m at 100m  – or sorry, if you actually are at 100m, what is actually reporting? Is it reporting 99m or 101m? What’s the range if it’s actually at 100? How truthful, essentially, is the CTD. (2) The resolution is: how much of a change can it detect? Again, think about pressure. Can you see surface changes if there’s a big tidal range in a particular area? Probably most CTDs will see that. But if there’s small surface changes, will that be seen by the CTD? (3) Then the stability. I think the stability is talked about the least, but I think it’s very important as well. It’s typically a unit over a period of time, which sounds jargony, but basically, it’s like X degrees Celsius over a year or X degrees Celsius over a month. Basically, it’s how you can trust the accuracy is going to be the same over that period of time. A good rule of thumb is that a good stability spec is the initial accuracy per year.


If it’s per month, that’s a little bit… it’s 12 times worse than having that accuracy per year. And that’s important because of the calibration piece. If you need an accuracy of one degree Celsius and the stability is half a degree Celsius per year, then now every year you’re expecting to lose 0.5 degrees of accuracy before it needs to be recalibrated. So really, the stability is an important piece for how often you need, or you should be, recalibrating.

What I’m talking about here in this last bit is the price. The price that you get on your quote is not the bottom-line price. There are other things that you should be thinking about, and the price is really a more holistic view of many different factors.

Now I’m going to talk explicitly about profiling CTDs, moored CTDs, and integration of CTDs at RBR. This is a very common CTD. It’s called the RBRconcerto3 C.T.D. This is the C, the conductivity cell, in this red here. The T is this, it’s called the thermistor. It measures temperature. It’s right here. And then this is the pressure sensor.

So how do CTDs capture this structure that I talked about earlier with the thermocline, the pycnocline, and the halocline? First, we have up to 32Hz sampling in the CTD, and the sensors have a fast time constant.


Between the three (conductivity, temperature, and depth; or conductivity, temperature, and pressure), the limiting factor in terms of time constant is the thermistor. We have a thermistor that has a 0.1 second time constant. Our standard thermistor is one second. This means that now when you’re profiling the water, the sensor is actually responding at 0.1 seconds. The other two sensors, conductivity and pressure, they’re essentially instantaneous. But like I said, the temperature is really the limiting factor there.

These are fast sampling. You can see here, if I point out this RBRconcerto, this is a thicker thermistor. And on this RBRconcerto over here, you can see it’s a thinner one. This is the fast thermistor and then this is the standard.

Back to the Wi-Fi piece: it doesn’t always have to be Wi-Fi, but for us it is that sanity check of checking the data while you’re out in the field. Did I capture that feature? We do have Wi-Fi enabled on all the CTDs. You need to make sure that, if you are going to buy one, tell your salesperson that you’re interested in Wi-Fi because it is something that needs to be turned on. And you can check the data on the spot, like I said. We have a free Ruskin mobile app. You can use the app or a computer to check it, it doesn’t really matter.


It’s a great way when you’re in the field to have that sanity check of “Did I capture the thermocline?” and check rough data. Using Ruskin, the desktop version, – you can download it right now, today, – and you can play around, without a need to do a whole bunch of different calculations, by simulating an instrument. And there is plenty of autonomy through batteries and the data piece.

I did run a quick check and for say, this instrument here, the RBRconcerto, if you did run it at 32Hz, it would be 19 days of autonomy. The memory would fill up after 19 days, and that’s if you’re sampling 24/7. So, every single day around the clock, 32 samples in every single second. That is quite a while.

So again, when you’re out in the field, if you’re going out for a day, and you have something like that, you don’t really have to worry about, “will I have enough memory or batteries to get through this deployment?”

Okay. So moored CTDs: looks very similar to the first one. I showed you a few differences. This is titanium, so it’s a deeper one. You can see this is a thicker thermistor. It’s meant for moorings. And we have some extra connectors back here. The deployment length for moored CTDs, again, I went to Ruskin and ran a simulation. And sampling once every six seconds on a RBRconcerto CTD, the battery will last about two years, just over two years.


So again, you can go into Ruskin and tweak around these factors to maximize – I didn’t do the bursting or the averaging I talked about, but you could also go in there and tweak it, if you know your deployment is going to be two years, then you can go in and tweak that until you get the right number of samples – or sorry, the right sampling interval – to maximize, you know, the metadata you can get on the battery and the memory.

The number of units: we actually have quantity discounts for when you buy multiple instruments. If you are setting up a mooring line that has many, many instruments, we have this awesome feature in Ruskin which essentially lets you plug in one instrument, configure it, and then without touching a button and just unplugging and plugging in, you can configure a whole series of instruments to deploy exactly the same way. Typically, on mooring lines, you want everything to be sampling at the same rate and the same time so that you can sync up the data later. Oh, and that feature is also really powerful for when you recover the instruments and you can automatically, as soon as Ruskin sees that you’re plugged in, download all the data and stop the deployment. That can be very efficient to set up and recover the instruments.


And then the power: I always say that power should never be a problem with your instrument because we have so many options for external power. It could be something like this where there’s a connector and you’re feeding in your own power. We have external battery packs like the RBRfermata that you can use to power the instruments. And then we also have an internal pack – it’s really doubling the length of the RBRconcerto by having this second housing that doubles the batteries. So, you’re really doubling the deployment. It’s not fully separate. There’s not a cable you can connect it on directly. Power should never be an issue for these sorts of deployments if you’re using an RBR CTD.

Okay, the OEM or the integration piece. These are two different variants of our glider CTD. We actually manufacture CTDs for many different float and glider manufacturers. We try to standardize it as much as possible, but they do have slight differences. For example, this is our RBRlegato, which is for gliders, and its pressure rating is down to 1000m or 1200m. If you need something deeper on your glider, you might need something like this, which is called the RBRcentauro. It’s better for deeper applications. We try to standardize a lot of pieces, but they’re all very much specified for the float manufacturers, and we have really great relationships with them to make sure that things fit size-wise.


The power draw from our CTDs is often multiple times lower in power than the status quo. That’s one thing to think about, to really extend missions on these autonomous platforms like gliders and floats. The power is a little bit shocking, I think, to people who don’t know instruments. Like I mentioned earlier, you get 19 days if you’re sampling around the clock at 32Hz with just a few batteries inside, essentially. These are super low power.

And the size: we actually designed these specifically with the size of these platforms in mind. For example, again, the RBRlegato  – there was a particular science bay that had to fit that in. It was built around that specification.

Okay. So, the physical considerations that were up earlier… Like the depth. We have CTD options down to 6000m. Our shallow version is around 750m. This is our shallow CTD here. I say shallow in quotes because for some people shallow is 20m or 50m or the shelf break. Again, you don’t want to take something that’s rated to 750 and try to see if it’ll go to 1000m. Please deploy your instruments to where they’re meant to go to! If you do have some rated to 6000m, if you try to deploy it in 100m of water… let’s talk about that a little bit.


It’ll still measure things. Maybe the pressure or the depth ratings won’t be as high resolution or high accuracy as you’d like. So, you know, really pick the right one. One question I ask is “how deep are you going?” or “what’s the deepest you’ll go?”

We have a lot of options for non-CTD parameters. You can get up to ten parameters on one instrument. It’s called the RBRmaestro. Outside of CTD, the most common things we’ve been asked to integrate are oxygen, chlorophyll a, and turbidity. There are many other things. Just ask us if we do integrations of specific parameters.

There’s a few of these sensors here. This one measures light and radiation. This is oxygen, and this is for turbidity and chlorophyll. And then you can get them integrated on here like this.

And then thinking about real time, or the near real time. We have options for a mooring line modem, like at very at the beginning of the talk, the folks that are deploying it through a hole on a mooring line that all the instruments are going to be frozen in. The data is coming up. The actual mooring line itself, it’s called a jacketed wire. So, the data gets transferred up there using modems. You can use external connectors and cables. We do have those options. Talk to your sales rep about what’s going to work best because there’s a limit to the length of the cable and how fast the data can be transferred and everything.


And then you can also use the Wi-Fi that I mentioned earlier as a data transfer piece. You can get the data off of the instrument without opening it up and download it and then share it on Dropbox or whatever.

And the environmental factors. I mentioned biofouling. For our optical sensors like these three at the top, we have wiper options. But for our CTD, hopefully later this year we’re going to be introducing an anti-fouling solution for our CTDs. And the second point here was about mooring line movement: if there’s a lot of currents and stuff like that, we have clamps that you can use for the line. You want to think about the dynamics of where you’re deploying it and how many clamps you need or how much backup tape you want to use. And CTDs are well known for operating well in cold climate conditions. So, if you freeze the CTD, it works just as well than when it’s thawed out. But again, the one thing to remember is that you have to consider the temperature piece. If it’s really cold, it’s going to drain the batteries a little bit more.


When you’re doing your planning, you want to maybe say half the battery life. So, it says 100 days, actually that’s 50 days because I’m deploying it in cold waters or in the Arctic.

Okay. So, these extra considerations I talked about as well. How good is the CTD, the accuracy, the resolution, and the stability? We have conductivity, temperature, and pressure, and then accuracy, resolution, and stability. I’m going to focus on pressure just for the sake of time. I’m not going to go through it all. But essentially, most manufacturers talk about pressure as full scale. This means the calibration of the pressure sensor.

I was earlier talking about the 100m pressure sensor. I’m going to stick there. If you have a pressure sensor go to 100m, that means for 0.05% that the accuracy is around 0.05m. If the CTD is exactly at 100m, the CTD from will give you a number somewhere between 99.95m and 100.05m. So very small, a very good accuracy to have that small of a change. And then the shallower the sensor, the better the accuracy, the better the resolution for the pressure sensors. The other things are standard across all pressure ranges. And then the resolution again, that’s how much surface change can I see? This is about a millimetre. And then the stability is the initial accuracy per year, as you can see.


So again, every year you can expect that your pressure measurement is going to be off by 0.05m if you’re not getting it calibrated. And because all of our stability specs are per year, that’s actually how often we recommend you get your CTD calibrated, to bring it back to that factory level the very first time that you get it. We recommend that. But also, you have to think about your science and what’s going to work best for you. So, you know, maybe you don’t need as high an accuracy and then you can get your CTD calibrated every two years because you’re okay with a drift after two years of 0.1m.

Speaking of calibration, or calibration turnaround, we have fast responses from our support team. And currently we’re doing something called a “three or free calibration guarantee” so you can get your CTD calibrated in three weeks. There’s some caveats, but that’s basically it, that you can get it done in three weeks. So that’s not months of having your instrument out of the water. If you are going to get your CTD calibrated soon, just let the team know if you have a tight timeline. And we often try to do what we can to really fit it in around your deployment. Of course, if you say I need it calibrated in two days, that might be a problem. So just give us lots of heads up.


The ease-of-use part: our CTD’s are literally shipped with twist activation enabled. And that sounds very weird, but essentially what it is, is the CTD, you get it, you open the box up, and you just twist the end cap and now it is collecting data at whatever sample rate you chose. It’s very, very easy to use. You don’t need to be, you know, a super expert in CTDs or an ocean tech or an engineer to use them. They’re very easy to use.

And the consumables. Again, those are the pieces that you have to change probably every time or, you know, every time you go in the field, you at least have to check them. There are three consumables on a CTD. The batteries, which are actually AA batteries. The desiccant, which is the thing that soaks up the moisture like little silica packets you get if you buy a new bag or a new purse. And the O-rings, which are included with the order, the spare ones. These are all user replaceable. You can buy the batteries anywhere. The desiccants are rechargeable, so you can bring them back to life yourself. And then the O-rings. You can always order more if need be.

Okay. I mentioned earlier that the quote price is not the bottom-line price. There’re other things to consider about the cost.


Of course, the cost of the instrument is a piece of this, but the cost is also how often you need to go and recover and swap instruments out to answer your scientific question. Right. So again, this is going back to “how good is the CTD.” You know, if the specs aren’t good, if you need something that has, you know, 0.5°C accuracy and it’s drifting by that amount every month, you have to go out every month to change it around. So having high quality specifications in terms of the accuracy and everything, that probably means you can leave your instruments in the water a little bit longer.

The cost of getting the instruments out of the water again. The calibration turnaround time so you can leave your instruments in the water for a year or two. That’s probably going to save you a lot of money for going out and doing recovery. The cost of a highly qualified personnel to operate. This is speaking to the ease-of-use piece. Is it really easy to set it up? How long does it take to learn how to use it? You know, are there lots of different moving parts and pieces that you have to kind of figure out? And the cost of the consumables. So again, if the consumables aren’t user replaceable, that means that the CTD has to go back to the factory. Or if the consumables are proprietary, that means that it has to be replaced there. How much is that going to cost?


Sorry. I’m just being conscious of the time because we only have a few minutes left. This is the summary. We’re getting pretty close! So, not all CTDs are ideal for the same applications. For profiling CTDs you really want to think about fast sensors with fast sampling. Ideally, it has a quick check feature, like an app or Wi-Fi to check, “did I get it?” The instrument needs to be able to handle the data, which is “will it be able to hold all the measurements? Will the batteries last while I’m out there for more deployments?” It sounds so silly to say it, but it’s true. You really need the CTD to last for the length of your deployment and to collect the data that you need to capture to answer the science question. Thinking about how easy it is going to be to set everything up, I love how easy it is to set up many, many instruments, to be able to easily deploy and recover. Again, power is a big thing that you have to think about for these sorts of deployments. And the integration piece. Make sure that the platform is right for you. That’s a good initial question. And you know, for the parts, or the sensors in general, you know that they’re the right size and the right power for the platform.


The physical considerations. Make sure you’re deploying to the correct depth. Think about if you need other non-CTD-parameters like chlorophyll or oxygen and how that might elevate the project that you’re working on. If you need real time or near real time, you need a connector or a way to send data to a surface and get the data off.

And then the environmental factors again, you want to think about biofouling and if it’s very dynamic, how are you going to ensure that the CTDs stay on there? And those extra considerations that we spent a bit of time on. How good is the CTD? How long does the calibration take? How easy is it to use, and how to deal with the consumables? Can you deal with it? Does the instrument have to go back to the factory?

Okay. I’m going to stop there. Thank you. If you feel more confused than ever, because that was a lot very fast, always write to us. This is not “I’m the only person who knows this.” Your sales rep will definitely know how to walk you through the options. I just wanted to give you some things to think about before you pick a CTD, with all the different components all together.


But again, just going back to the application, how are you trying to put the CTD in the water? That will answer a lot of our questions to get you down the right road, to get exactly what you need to make sure that your deployment is successful.

Okay. I’m going to check the chat. Okay. Okay, great. So, a question that I mentioned earlier is pretty common is the maintenance piece. How much maintenance does an RBR CTD require? There’s really two pieces. I think the first is the general maintenance for when you’re going out. So those three things I mentioned earlier, those consumables, the battery, the desiccant, and the O-rings. Every time you deploy, you should check all three. Because the batteries, if you don’t have enough juice, it’s not going to last for your deployment. The O-rings are super important because if they leak, your instruments are destroyed, and then the desiccant actually keeps the inside of your logger dry so that when you close it up, and there’s condensation inside, it soaks it up. If it’s spent, you could get corrosion inside and then that will actually affect your battery life. So those three things, the O-rings, the desiccant, the batteries, every time you deploy, you should check those things. That’s one maintenance piece. And the second thing I would say is get your CTD calibrated every year.


That’s the other piece. Because when we get your CTD in house, it’s not just sent exactly to the lab and that’s it. Our service technicians actually take it and do a whole process of checks on the instrument to make sure that everything is up to snuff. So, they check things like the sleep current. If there was corrosion that you didn’t see, now they’re checking it. Oh, there’s actually corrosion on this logger. They haven’t been changing their desiccant. So now when you get the instrument back, support will reach out to make sure to tell you what’s going on and say, okay, you have corrosion, we need to fix this or replace it. Other than that, there’s no main maintenance thing. Keep your CTD clean.

I mentioned Biofouling earlier. The conductivity cell is susceptible to biofouling, but you can go out and service it and keep it clean. But with the battery life being what it is and the memory being what it is, there’s not a lot of hardcore maintenance to it. If you are cleaning the CTD, we have suggestions on our docs site that will tell you how. (We can send the link to that.) It just tells you, if you’re doing this, if there’s a lot of growth on there, you can use a vinegar bath, and such things.


But these are very robust sensors, so there’s not a lot of maintenance. To circle back and answer the question, check the consumables, which are the batteries, the O-rings, and the desiccant before every deployment and replace as needed, and get your CTD calibrated every year. Because in addition to getting the very best quality data again now you’re also getting this in-house sanity check, “is the CTD working right?” So that’s what I would recommend.

I’m just looking at the other questions. Okay Kelsey is asking what determines how quickly a profiler needs to go through the water. So – and Kelsey, you can just hop in if I’m answering the question right – I think there are two things to think about when you’re profiling. One is, like I mentioned, the actual response time – not the response time, sorry, how fast the sensors are taking data. We can do 32Hz. That’s one thing. The second thing is how fast the CTD is going through the water, so typically, when instruments are in free fall. Our instruments, depending on what kind of instrument it is, the smaller ones, there’s less weight, they fall through the water more slowly. If you have a cage that’s going to add weight, it’ll move a bit faster. Typically, I’m guessing you’ll see a fall rate between 0.5m/s and 1m/s.


If you’re falling at 0.5m/s and you want, you know, 16 data points every metre, then you can sample at 8Hz. If you’re going 1m/s, you need to get those same data points, it’s 16Hz. So really it comes down to your scientific question. If you have a very, very sharp thermocline that you’re trying to see, you’re going to want to sample as fast as you can through that and go relatively slow through that. If the thermocline is changing very quickly – and some places have tanks set up that, if it’s super, super stratified, you can test that out a little bit at some facilities. So those are the two things.

Now, the winch part…Later this year we are releasing a winch here at RBR that spits out the line so that you can measure while the boat is moving. That’s another piece, if the boat is moving really fast, the line needs to be spitting out really fast. If the boat’s moving slower, you have to spit out the line a little bit slower. So, it’s a combination of a few things, but typically: how fast the CTD can sample, and how fast the CTD is falling through the water. Those are two things you really want to think about, and then that will answer your question of how fast does the line need to spit out? And if you have a boat that’s moving, that’s a whole other piece in there.


But again, it’s those pieces. But I would strongly recommend you go back to your science question – I don’t know about you, but when I was in grad school, the region that I studied was already well studied, so we knew what to expect. So, if there was a really sharp thermocline and we were seeing it, we knew what level of CTD sampling we needed. I hope that answered your question, Kelsey.

How have we studied the freefall speed of our CTDs through the water? If so, is there an average number? I think those ballpark figures, I gave 0.5m/s or 1m/s, are common, but we haven’t because people are using them all different ways. Earlier, I showed a picture of our research scientist Mat Dever, and he had a winch and you could see him holding it out. You could see a little bit of the CTD at the bottom. But he, during his time at WHOI, helped develop this collar that goes around to make the CTD fall through the water faster, up to 4m/s, underway, so you can add weight. Again, reach out to us about how that’s best done.


Because there’s just some proximity effects around the conductivity cell that need to be considered. So, if you need faster or slower, we can give you some recommendations for that. So, Kelsey is asking again about the profiling speeds. I should have mentioned this. There’s just so much information, sorry. So typically, when you’re profiling, let’s say you’re using a winch, so you’re profiling the CTD down and you’re pulling it back up. The undisturbed water is actually on the way down. So, we face our CTDs down, essentially. And then as you’re pulling it back up, usually people don’t use that pulling back up data because the CTD sensors are all there and then you’re pulling it through the water so you’re disturbing all of the structure. So, if you are measuring a thermocline, it might be really hard to see it on the way back up. Whereas if it’s going down, it’s measuring this undisturbed water, so you’re getting all those features. Typically, the payout speed is more important to get it down, to get a really nice profile. But when Mat’s been testing our new winch, he loves the speed on the way back because it’s saving so much time. So, if you have a really slow payback speed, it can be kind of a pain because you’re losing precious sampling time.


Our winch will be paying out and paying in at similar rates. I don’t know if it’s exactly the same rate, of course, because there’s safety things you don’t want, like the thing zooming back in. I would say the direction that the most undisturbed water is the most important for the sampling rate. On the Wirewalker, actually the way it works, it ratchets down and a lever is released and then it profiles upward. It has a clean profile on the way up. The fast sampling is important for that. Again, there’s so many pieces to all this, but the Wirewalker or our CTDs actually have this cool feature where you can profile fast in one direction and slower in another direction, which is really cool.

There’s a question about thermal shielding. If the instrument was in sunlight for a while, say on the deck, and then you’re putting it in cold water. I have recommendations for this. So first of all, let’s say you’re on a typical beautiful summer for your cruise, so you’re going to put the CTDs on the deck, and it’s in the sunlight. Now, the conductivity cell is 25°C and you put it in the water and the surface water is 20°C.


We recommend a little bit of a soak at the surface to get that thermal mass, because as you can see here, this is a, you know, a piece of plastic. So, it has to heat up or cool down. So, if there’s a big temperature difference, then you want to leave it in the water for the you know, for the CTD to heat up or cool down. Some of our customers in the Arctic do a really cool thing, because as you can imagine up there, the air temperature can be very, very cold, like -25°C. And then the water is going to be at the coldest -5°C because that’s around where saltwater freezes. So, for that, that’s a huge jump. So, then I recommend people getting a cooler, like a cooler that you would take to the beach in the summer, and you put your CTD in there with 0°C water, like cold-to-us water. But then you can put it in there and then keep it closer to the temperature, so you have less of a time in the cast. The warm body will affect the temperature. So again, if you soak it at the surface, in time, the CTD will acclimatize, and the temperature will be cooled down or heated up as much as it needs.


For the upcast, it depends. If you’re using it on a Wirewalker, on the upcast the sensors are facing up. I don’t know if people see much effect if the body’s a little bit warmer and you’re pulling it through. I think that the data quality piece, if you’re profiling down and then pulling it back up, people aren’t really using that data on the way back up because it’s not measuring undisturbed water.

Whoops. End of slideshow. I don’t see any more questions. Keith is going to launch the poll now. I know, it’s really close. It’s on the hour. There we go. You can answer the question. Sorry, we’re right on the hour. If you have more questions, I’m going to move this poll out of the way a little bit, you can email My direct email is my first name dot my last name at Feel free to take a screenshot. There’s a couple of poll questions there, which would be great if you can answer. And again, we will be in touch with the link to this and some of the Q&A answers and that cool infographic. I cannot vote for this poll. That’s not fun.


I mean. It is kind of fun still. I mean, I’m glad you are answering, but it would be fun for me to see the coolest way a CTD has been deployed. I’ve had people ask too about – and I almost put it in, but I actually haven’t heard somebody do it – from drones and dipping it into the surface. That might be interesting to try to do that, to get surface stuff. But depending on the size of the drone, it might be relatively heavy. Even though RBR CTDs are pretty small like this one, the RBRbrevio. I think also when you answer these questions, if there’s something really cool, we can throw that into the follow up email as well. There might be other things. And again, if you have other thoughts about CTDs, feel free to throw it in the chat. Oh, awesome. Okay, so if you’re still here, we are sharing the poll answers. I think you should be able to see that on your screen. It looks like most people go to the vendor websites to decide on selecting instruments. That’s really interesting. Cool. And the reason I’m saying that’s interesting is because I often think that you ask your colleagues. That’s cool. And then price often seems to be the most important thing when selecting a CTD. Okay. I’m going to call it. Sorry that we’re three minutes over. Maybe next time I should start on time will be three minutes over. Thank you all for attending again. Contact me. We’ll be in touch.


Read more user stories


Date published 2023-07-14