TCarta Dives Underwater to Provide Satellite-Derived Bathymetry

Sanjay Gangal, CEO and president of GISCafe, interviewed Ross Smith, program manager at TCarta recently, to talk about the importance of TCarta and its satellite-derived bathymetry, or SDB to climate change, sea level rise and GEOINT.

TCarta is a hydrographic and hydrospatial remote sensing company that specializes in the marine domain using space-based sensors for C4 mapping, C4 specification and primarily, satellite-derived estimation mapping and satellite-derived bathymetry, or SDB.

What is Project Trident? asks Gangal.

“Project Trident is a National Science Foundation Small Business Innovation Research grant,” says Smith. “We’re currently in phase two, and the whole scope of the project is to really try and bring satellite-derived bathymetry into the modern world with machine learning, deep learning and some level of automation and taking a lot of the common detractors and difficulties of satellite-derived bathymetry and turn them into technical solutions. The primary focus is using multiple space-based sensors in order to derive a global bathymetry model of the shallow waters around the world. So it’s entering the end of the third year now and we’re very excited about it.”

“Okay, great,” says Gangal. “So terrestrial digital elevation model products from satellite imagery are easily available and automated global products exist but not for satellite-derived bathymetry. What makes SDB such a challenge and what has been the TCarta’s approach to address it?”

“Working under the water is definitely an added level of difficulty. There are many factors to that, primarily being that most images aren’t going to be suitable,” says Smith. “You have a lot of difficulty with water conditions and water service conditions, as well as atmospheric conditions that make selection and assessment of imagery typically a very difficult and analyst-driven process. You also have the challenge of accessing imagery in volume and at scale to accomplish something like a global bathymetry model. And then you also have the challenge of available in situ data in order to calibrate or validate your processes. One of the main reasons why we’re trying to create a global shallow water bathymetry model is because there’s not a lot of data in a lot of remote places in the world where it’s needed the most, and so those are probably the largest challenges and the ones we’re trying to overcome with our Project Trident.”

How has deep learning, AI or machine learning change the way TCarta produces data?, Gangal asks.

“A lot, quite frankly,” says Smith. “One of the primary sensors that we’re using currently is the ICESat-2 altimeter sensor. It’s an active space-based LIDAR sensor from NASA and we’ve been on the applied user program since the launch, effectively. The sensor has a 532-nanometer wavelength laser. It’s a blue-green laser which is perfect for mapping shallow waters. However, the mission of the satellite was never actually for shallow water bathymetry mapping, so we were able to manually produce the data initially and then realized that it was a very time-intensive process and that it was a great opportunity to leverage machine learning and deep learning to accomplish that. And we have definitely been able to use this as the point, effectively a space-based in situ source, talking about the challenges we’ve had and the challenges of SDB with not having calibration and validation data. With this space-based LIDAR sensor, we can follow ourselves around the world and using this deep learning, create a very accurate sub 50-centimeter, 30-centimeter accuracies to then inform our imagery for the continuous surface.

I believe, to date, we have over 140 million ICESat-2 bathymetry points derived on the coastline of six continents, so that’s been a huge force multiplier. And then we’re also using machine learning and deep learning for segmentation and post-processing of our SDB surfaces, so the continuous surface you get from a full image of, say, Sentinel-2 multispectral imagery, most of that data is either going to be on land or deep water, and then you have your shallow water, which is what we want. So again, a very analyst-intensive process. Typically, we were able to utilize machine learning and deep learning in order to automate that process or at least partially automate that process.”

“How have your users utilized satellite-derived bathymetry and what benefits to the reporter do you see?” 

“There are many benefits to satellite-derived bathymetry,” asserts Smith. “We have a lot of hydrodynamic modeling, and a lot of coastal engineering projects, as well as environmental resiliency projects and monitoring, coastal change monitoring. You can even extend it into, most recently, nautical charting. There have been some major hydrographic offices that have incorporated SDB, including some of TCarta’s SDB. And then you also have survey planning. So if you want to do a more pointed, more traditional LIDAR or multibeam echosounder survey, SDB offers a great way to scope out the area and plan your survey. So there are many, many applications and I think our users find that the affordability and the speed at which you can produce it is quite the biggest benefit.”

“And how can TCarta’s technologies be used for GEOINT applications?”

“That’s a big one. Any time you have a location where you might not be able to put a boat on the water for various political or restrictive reasons, having a completely space-based remote solution is very powerful. And so say you have an island being built somewhere where maybe it shouldn’t be and you can’t put a boat on the water, and the charts are from before the island was built. Having the capability to measure the depths with two different sensors and corroborate that information to get a new depth model provides a very powerful tool for operational planning, beach landings and things of that nature. It’s also of great benefit to natural disaster relief response, which is also a GEOINT application as well.”

How can end users access TCarta technologies or products? Asks Gangal.

“We have two models from Project Trident. We have the data-as-a-service and we have the software-as-a-service. So the data-as-a-service obviously is going to be the data we are producing for this global bathymetry model, and that is persistently updated and the standard at a 10-meter resolution from Sentinel-2,” notes Smith. “We have actually already started producing higher resolution 2-meter SDB products as well, and those are also available. And then we have the software-as-a-service, which is basically just trying to expose some of the technology we’ve developed while we’ve been on this research program and giving that to end users as well. So if somebody wants to not use their own imagery or has a very pointed use case, they can use our software-as-a-service, and that comes in the form as an ArcGIS Pro geoprocessing toolbox, and that includes some data preparations, some data analysis tools, as well as two different mechanisms for satellite-derived bathymetry processing.”

“Now with regards to climate change and sea level rise, how can satellite bathymetry help inform better decision-making for coastal resilience?” Gangal asks.

“That’s another big one,” says Smith. “A lot of that is going to be reliant on some of the similar applications, the broader ones, i.e., hydrodynamic modeling related to sea level rise or to storm mitigation and also, storm surge and even tsunamis. You can really use that information, especially corroborated with a terrestrial BM, which is known as a topobathy model or topographic bathymetry model. Those two data sets together really corroborative for doing this, like a sea level rise monitoring or storm mitigation. And you can also do change detection as well, so how has the seafloor changed with different weather patterns or how tide rises of sea level can affect the bathymetry as well.”

“What’s the differentiator between TCarta’s approach to satellite-derived bathymetry compared to other vendors or researchers?” Asks Gangal.

“There’s a lot of great research and a lot of really, really great things being done in the realm of remote sensing around the world,” Smith explains. “I think the thing that TCarta does the best is we really don’t try to stick to one game plan or one method, we’ve really recognized that the world is very dynamic, and especially in the marine domain, very dynamic, and not one tool is going to function in one location. And so we’ve really taken the approach of applying a multimodal approach of different satellite-derived bathymetry modeling techniques in order to ensure that we’re getting the most accurate data possible in a given location, and using the right tool for the right job. I think that is probably our biggest innovation in addition to the deep learning applications and ICESat-2, which are the best in the industry.”

In order for the GISCafe readers to find out more about TCarta solutions and TCarta on the internet, go to tcarta.com. There is also an email: info@tcarta.com as well as contact forms on the website.

Tags: climate change, data, geospatial, GIS, health, imagery, Infrastructure, intelligence, LiDAR, mapping, navigation, remote sensing, satellite imagery

Categories: analytics, Big Data, Building Information Modeling, climate change, cloud, data, developers, disaster relief, earthquakes, election maps, emergency response, geospatial, GIS, handhelds, laser radar, lidar, location based services, location intelligence, public safety, satellite based tracking, satellite imagery, sensors, spatial data, storm surge, subsurface utilities, survey, transportation, utilities, Video Interview

This entry was posted on Thursday, October 14th, 2021 at 9:08 am. You can follow any responses to this entry through the RSS 2.0 feed. You can leave a response, or trackback from your own site.