This is Overwatch, a podcast presented by the Institute for the Study of War. ISW is marking the first anniversary of Russia's 2022 invasion of Ukraine with a three-part interview series between ISW Russia team lead Mason Clark and senior members of his team.
In this episode, Mason and Russia analyst and geospatial team lead George Barros will discuss the impact of HIMARS and other advanced Western weapons, as well as the nature of Vladimir Putin's so-called "red lines" in the conflict. George will also provide insight into how ISW produces its world-renowned maps of the war in Ukraine.
I'm Mason Clark, Senior Analyst and Russia Team Lead here at ISW. Today, I'm speaking with George Baros, Russia Analyst and Geospatial Team Lead at ISW. George has been with us since fall of 2019 and has covered the Russian military and also focused on Belarus for much of that time.
He led a lot of our coverage of the Russian buildup to the renewed invasion of Ukraine in February 2022 and is responsible for the wonderful control of terrain maps that you've seen on ISW's reports as well as in all of the world's leading media outlets.
George, thanks for joining me today. Thank you so much, Mason. So for this series, I'd like to talk to each of the senior members of the team about their key insights on the war to date around the one-year anniversary of the Russian invasion on February 24th, as well as dive into the methodology and analysis that they use to inform our reports and where they see the war going in the coming year.
So, George, starting off with the general question, what are some of the key inflections and important themes of both Russian and Ukrainian capabilities in the first year of the war that you focused on? Thank you, Mason. It's been very fascinating to watch and see exactly how the Ukrainians have been such very good masters of operational planning throughout this war and this campaign.
Every single time that the West has provided Ukraine new weapons systems that grant them new operational capabilities that they previously didn't have, the Ukrainians have time and time again demonstrated that they can integrate them to great effect and actually achieve significant battlefield gains. So what do you think are some of the most important Western systems that have been provided and then how have the Ukrainians used them?
Sure. I'd like to focus on the HIMARS, the high mobility rocket artillery systems that the United States provided Ukraine back in the summer. This was a really critical system that enabled the Ukrainians to hit targets in Russian rear areas that the Ukrainians previously didn't have any equipment in their arsenal to be able to hit.
The United States first gave Ukraine these systems back in June, and it actually helped with culminating the Russian offensive in eastern Ukraine and set really good conditions for Ukrainians to conduct the counteroffensives that they did over the summer and the fall.
So I remember covering that in a lot of detail of combining the different sources about Ukrainian strikes on Russian rear areas, both in southern Ukraine and in eastern Ukraine, pulling together all of the different sourcing from Russian mill bloggers, Ukrainian statements on their strikes. But I'd like you to talk a little bit about what you focused on is the visual component. Could you talk to us about how we've assessed those Ukrainian strikes and how they've impacted Russian operations?
Sure. So we collect a tremendous amount of data, as our readers know. But really, I think one of the value adds that our analysis benefits from is contextualizing the data and looking and seeing physically where these videos and where these textual locations that are mentioned in our sources actually physically are on terrain, on the map, and looking at how it impacts battlefield geometry and the ability for the Russians to conduct their operations.
When the Ukrainians received the HIMARS, we started to see the Ukrainians conduct these very targeted, deliberate strike interdiction campaigns against key Russian supply lines and against key rear logistics nodes.
ammunition depots, headquarters, concentration areas for forces and equipment, garrisons, things of that nature. And actually due to the way that the Russians wage their war and largely leverage high volumes of artillery fire, that necessitates the Russians to bring a whole lot of ammunition very close to the front lines.
And when the Ukrainians could actually strike those, it degraded the Russians' ability to, number one, mass all that artillery fire and then actually use it to destroy Ukrainian positions to advance.
Right. And that's been a very, the Russians have had to employ a very static type of fighting throughout a lot of the fighting, particularly around Severodonetsk last summer, and then the ongoing fighting around Bakhmut and other areas in eastern Ukraine. Now, one of the most fascinating elements of this, personally, for me, when we've been unpacking the daily updates, I've very much focused on the writing and have watched you do a lot of the controlled terrain changes. And it's always been, you know, of
I'm a little biased, but it's been fascinating watching how the mapping team breaks down all of the scattered reports, especially when the changes are so minor. So I was wondering if you could walk our listeners through sort of an example of how you break down the small and very conflicting reports of terrain changes and how that informs the overall daily maps. Sure.
There's three big sources that we use in order to create the control of terrain. And those are visual sources, textual sources, and then data that we remotely sense. That is data we collect with airborne or spaceborne sensors. Visual sources are very straightforward. We see a video of some combat occurring somewhere, or we see a selfie that Russian or Ukrainian forces took. We can geolocate that, or a larger open source community can geolocate that.
and we can figure out where physically on the Earth that occurred. And if we can determine where in space and in time that activity, that data comes from, that becomes a mappable, plottable data point on the map. So, sorry to interrupt briefly, but for those that aren't familiar, could you describe in a little bit more detail what you mean by geolocating? How do you figure out where something is in space? Sure. Geolocation is a process of determining locations
where physically in the world a particular piece of evidence occurred. So for example, in the context of this war, if we see Russian forces at Hostomel Airport right outside of Kyiv, we know we can determine that it actually indeed occurred at that particular place in time because maybe we'll see some unique
features featured in the video or in the photo that we can actually compare and contrast against previous historical known reference images of that location. So for example, to do this in a different kind of context would be like, let's say you go to the Epcot Center and you take a picture with your family there and there's a picture of that big white ball.
And you post it online and you don't provide any context about where this location is. But because people can study this picture, they can see sort of the unique visual phenomena that are in the picture and they can deconstruct it and say, ah, this picture is actually located here at this particular place at the Epcot Center because there's all these unique buildings here that don't exist anywhere else. And we can look at historical images to compare and contrast and confirm, aha, this person was at this place at this time.
Right, and that's why it's always fascinating looking for the sort of background buildings and terrain features in videos and images we look at. You know, it's always very helpful when you see a building with a roof with a strange color of like, oh, there's that building with the bright red roof in Soledar. OK, they must be from this hill, that sort of thing to unpack where exactly footage is coming from. That's only one aspect of what goes into the daily control terrain maps. Yeah, so those are visual sources and they're fairly straightforward.
The other major source that we exploit are textual sources. These are not visual, but just simply written prose describing condition of the battlefield, condition of various cities and towns. These come from everything from the Ukrainian general staff daily reports to Ukrainian oblast level, provincial level authorities, sort of their county equivalent authorities, social media users that are there.
It also comes from the Russians. It comes from the Russian Ministry of Defense's daily report. They come from Russian military units and what they read out, as well as from Russian military bloggers and war correspondents who oftentimes are embedded with those units. And it's fascinating is oftentimes they will provide very detailed accounts down to the street level of where particular fighting is occurring. Watching the fighting in Bakhmut and Mariupol and in other urban locations has been very interesting because
These conflicts occur for many months at a time, and they will describe Russian military bloggers down to the street level. Today, we had made advances on Horobovsky Street, or we made it down like Vatulin Avenue or something.
And they will describe exactly how many blocks they can advance, what streets are contested. And if you actually aggregate and systematize the way that you collect this information over time, you can watch the block by block fighting and clearing of an urban environments, which has been interesting to visualize.
Right. And so the final element we use, you mentioned, is remote sensing, which is both satellite imagery and a couple of other things that we've used for some cool products like looking at Russian defensive positions and stuff like that. What are some of the more creative ways that you've used those remote sensing opportunities to inform the maps? So just briefly for listeners who might not be familiar, remote sensing refers to the activity and science of collecting data from a place where you physically are not or cannot be.
So we obviously can't go to the front lines in Ukraine. So sometimes we have to be able to rely on other means of collecting data that we can't just go there ourselves.
So we rely on a lot of spaceborne sensors and satellite imagery in order to help build out our picture. So we'll collect your standard Maxar images. There was a really excellent piece that our researcher, Grace Mappis, wrote back in late November that talked about how we got a bunch of images from Maxar that showed where the Russian field fortifications at Kherson is. And based off of those data points, we could determine that or rather assess that the Russians had established a rather
intricate layered line defense in Kherson, which indicates that the Russians take seriously the threat of Ukrainian advance across that river. We also leverage other forms of remote sensing though, beyond just sort of your traditional satellite imagery.
We take in, for example, radar data. One of the big caveats and issues with working with traditional electro-optical imagery is that cloud cover really affects your ability to collect on the ground. If you were to point your spaceborne camera at the Earth and there's a bunch of clouds in the way, you're just going to see the tops of clouds. You're not going to see the tanks or the other things that you want to see.
But radar is another great technology that will penetrate through the clouds and create a sort of a 3D radar image of whatever's there on the Earth, because those particle beams just go from the sensor, through the clouds, to the Earth, back up. So that's a great thing that we're able to collect on. Another great thing that we're able to collect on as well is different types of hyperspectral data. So not images, but frequencies and signatures. There's one source that I really like to exploit in particular, which is the NASA FIRM sensor.
Right, and that's actually originally a tool used to track forest fires that has been remarkably effective for tracking Russian and Ukrainian artillery, areas that have been struck, fires in towns, a number of other items like that.
Now, we could nerd out about what goes into the maps and the updates all day, but I want to zoom back up a level and talk about what you're observing of the sort of status of the war as we approach the one-year anniversary of the invasion on February 24th, and where do you see it going? And George, I know a lot of your work in the last year has focused on sort of myth-busting, in a way, the coverage of the war that's been a huge contribution to our updates. So what are some misconceptions you're still seeing about the conflict and where it's going?
Sure thing. I think one of the retrospectively, not forecast, but retrospectively, one of the big myths that we've seen is that the Ukrainians actually can't conduct offensive operations. They cannot conduct maneuver warfare and that they're always going to be stalled out. I remember back in the summer, there were some really intense debates about whether or not the HIMARS we sent them made a difference.
and will the Ukrainians actually get their Kherson counter-offensive off the ground? Because when they announced it back in August, there was actually not a whole lot of movement for several months.
And I think it's very plain to say that the data shows that that myth is thoroughly busted. Remote sensing data actually supports that. So right after the Ukrainians received the HIMARS and they actually began striking all of these ammunition depots that were relatively close to the front lines, but just outside of the reach of the previous Ukrainian strike ranges,
we started to see a massive drop off for these detected heat anomalies by the NASA firm sensor in Eastern Ukraine. And that occurred around July. There's this very distinct pattern in the historical data where every single day we see a whole bunch of heat anomalies and we see a whole bunch of visible infrared spectrum signatures consistent with fires from Russian artillery. But then the Ukrainian HIMARS come in, they
They start striking the depots and that number starts to drop off. And then actually, because the Russians lose that ability to mass artillery, their campaign stalls out and culminates in early July, right after they capture Lysachansk. The HIMARS artillery also enabled not only the culmination of the Russian offensive, but enabled Ukrainian counteroffensives in the south and in Kharkiv.
So you've also looked a lot at the sort of information space dynamics of these Western deliveries and how the Russians have or really have not responded to those. What stuck out to you the most about the sort of gap between the Kremlin's claims about how it will respond to Western deliveries and then what's actually been borne out in the last year?
Yeah. It's been very interesting to see how the Kremlin demands and its articulation of its so-called red lines have evolved really over the past year. I know we're talking about the anniversary for the war going back to last February 2022, but –
We've been tracking this for far beyond that. And I remember, you remember very well, late 2021 watching the weird stuff with Wendy Sherman and Lavrov and all those weird meetings in Switzerland and what have you and Geneva about the Russian security demands and security guarantees and all these so-called red lines. Literally, not journalists in the West calling them red lines. Putin literally and Lavrov literally saying in their speeches, these are red lines. Things like there could be no more NATO expansion. NATO has to go back to its 1994 borders.
You can't have any weapon systems that can strike the Russian mainland. And you certainly may not actually violate the sovereignty of the Russian mainland and you may not strike targets in them. What's been really interesting is now, after almost a year of war, Ukrainians have actually contested every single one of those red lines, violated them in not obscure terms. And the Russian response, where they promise drastic but unspecified retaliation, has never actually occurred.
And I think therefore, the data-driven key takeaway is that Putin's red lines actually really aren't that real. It's actually probably designed to constrain Western decision-making and constrain what we're willing to do out of fear for what the Russians might actually do.
Right. And you've certainly written and talked extensively about how important it is to provide Ukraine with the systems it needs to then carry out that, as you mentioned earlier, very effective operational planning that has enabled those past counteroffensives in Kherson and Kharkiv, as well as preventing Russian successes.
Now, we could keep going about this for hours and hours, but before we go, do you have any top line of forecasts of what are you watching related to this provision of Western aid and Russian responses going into the rest of 2023? Yeah, the key takeaway is the West, if we're actually serious about we want the Ukrainians to win and be successful, then we need to take the gloves off and actually provide the Ukrainians the systems and the tools and the sustainment over time so that they actually can win.
It's been fascinating to watch this Western aid dynamic over the span of many years. I remember back in 2014 when the United States government started first talking about sending Ukraine basic, really bare-bones security assistance. It was extremely controversial, the idea of sending the Ukrainians sniper rifles and a small quantity of javelins.
We finally convinced ourselves in 2016, 2017 that we can send them a small quantity of javelins, but they have to be garrisoned in warehouses in Western Ukraine. They can't go close to the front. And then in 2019, we said, okay, javelins can go closer to the front. And then now we've just been seeing sort of the escalation of this exact same pattern where we say, oh, no, we can't send them
field artillery. We actually can. We can't send them HIMARS. We actually can't. We can't send them in battle tanks. Oh, actually we can. And now we're having this debate about fixed wing aircraft. We need to really stop this whole charade. We're keeping Ukraine on a starvation diet for the aid that it needs to actually win. And the cool thing about the Ukrainians is that they're very responsible recipients of this military aid. Every time we've given them these systems,
They've excelled and actually been able to integrate them into their operational planning quite successfully. When we gave the Ukrainians the HIMARS in June, they very promptly started their interdiction campaign over the Kherson bridges in July and in August. And they shaped the conditions for the battlefield to begin their maneuver phase in late August. And that continued all the way until November 11th, which is when the Russians, because their ground lines of communication were so degraded, they pulled out.
Frankly, the key takeaway is the Ukrainians probably could have liberated Kherson a lot sooner if they had received the HIMARS sooner than they did in June. So that's the key takeaway for sort of the retrospective for what we've learned about the conflict so far.
Looking forward into the future for our near-term forecast, over the course of the winter, we were forecasting that the Russians would likely begin their new decisive offensive effort within the next six months. We now assess that the Russians are starting to slow roll that offensive, beginning that in late January, early February. I think the key narrative and the key misconception and myth that this busts is that stalemate is inevitable. There's been a lot of talk about how, you know,
We've recently seen a lot of positional warfare. The Russians are exhausted. The Ukrainians are also degraded and exhausted, and therefore, nothing is going to really happen. It's going to stalemate, and why would we consider supporting what's going to inevitably be a protracted war of attrition where there's going to really not be a whole lot of maneuver or decisive action? What the Russians are doing now is it demonstrates that actually, no, this is still going to be a hot war moving forward for Ukraine.
Certainly this next year, 2023, stalemate is not inevitable. And when the Ukrainians finally get to ingest all these main battle tanks that the West has just recently pledged, that's going to set good conditions for the Ukrainians to once again try to conduct some counteroffensives to liberate significant portions of their territory. Certainly. Well, I know you'll have your hands busy with continuing our daily maps of the war for quite some time, but hopefully going to be adding a lot more of that light blue Ukrainian counteroffensive section to the map.
Well, George, thank you so much for speaking with me today about some of what goes into those daily control terrain maps and some of the dynamics you've been observing and writing about with Western Aid. Thank you so much, Mason. It's a real pleasure. Thank you for listening to this episode of Overwatch. We look forward to your feedback on this episode and previous ones. Visit www.understandingwar.org to learn about ISW's work and to sign up for our mailing list.