Latest / Stage Talks with Bellingcat / "Studying Conflict’s Impact on the Environment" with the Conflict and Environment Observatory
Transcript
- 0:04You're listening to a stage talk titled Study
- 0:07and Conflicts Impact on the Environment. This
- 0:09week we were joined by Dr Owen Derbyshire from
- 0:12the UK charity, the Conflict and Environment
- 0:14Observatory. Owen spoke about the many different
- 0:17types of conflict pollution and environmental
- 0:19harm they've come across and how the organisation
- 0:22works to monitor it, sometimes for years after
- 0:25the initial fighting has stopped. You can find
- 0:27links to all the resources mentioned in the talk
- 0:30in the podcast description. This talk was hosted
- 0:32by me, Charlotte Ma, on Thursday the 20th of
- 0:35November 2025 in the Bellingcat Discord server.
- 0:49Hello and welcome back to the Stage Talks series.
- 0:52This week we're exploring the impact of conflict
- 0:55on the environment. We all know the damaging
- 0:57effect of war on society, but the often overlooked
- 1:00toll of war is on the environment. Someone who
- 1:03works to cover this harm is Dr Owen Derbyshire,
- 1:06an environmental scientist working at the Conflict
- 1:09Environment Observatory. The Conflict Environment
- 1:11Observatory is a UK charity with a mission of
- 1:13increasing awareness and understanding of the
- 1:15environmental and derived humanitarian consequences
- 1:18of conflicts and military activities. They also
- 1:21work with organisations to develop tools to improve
- 1:24data collection and sharing. Their reports focus
- 1:27on regions around the globe with their latest
- 1:29investigating ongoing conflicts from Ukraine
- 1:31to Sudan. They're also about to release a new
- 1:35database showing the incidents in a searchable
- 1:37format. Hopefully by the end of this talk, you'll
- 1:40have an understanding of what can be measured
- 1:41when it comes to environmental damage caused
- 1:43by conflict and how those measurements could
- 1:45be used to enact change. As we talk, please make
- 1:49sure to add your questions in the chat box via
- 1:51the message bubble icon in the top right corner
- 1:53of your screen. and please note within your question
- 1:56if you do not want me to read your username out.
- 2:00Okay, Owen, I'm going to mute my mic and pass
- 2:04it over to you. So yeah, hello everyone and thanks
- 2:07for having me. So environmental scientists, we
- 2:11often struggle to document harm and conflict
- 2:14zones. We can't get to sites, samples are dangerous
- 2:19to collect and information is tightly controlled.
- 2:23But with OSINT, we have opened doors that were
- 2:27previously shut. So we now have satellite archives,
- 2:30we have underground footage, and we have community
- 2:32reporting. And so what we do is bring together
- 2:35environmental science and OSINT. And this allows
- 2:39us to understand wartime environmental harm in
- 2:42ways that simply were not possible a decade ago.
- 2:47But first of all, let's start with a big question.
- 2:50Why? Should we care about war and the environment
- 2:53given everything else that happens in war? Well,
- 2:58environmental harm and conflict translates directly
- 3:00into harm to people. Pollution, destroyed ecosystems
- 3:04and collapsing services create immediate health
- 3:07risks and they also leave long shadows. You know,
- 3:11sometimes for generations, for example, the legacy
- 3:14of herbicides in South East Asia. Just one example
- 3:17of how long these effects can linger. And once
- 3:20the war finishes, once the front lines move on,
- 3:23people remain and they're left to live amongst
- 3:27toxic residues, damaged infrastructure, chronic
- 3:30exposures to pollution and maybe more importantly
- 3:33governance vacuums where further environmental
- 3:37exploitation can flourish. We should also care
- 3:40because the environment has intrinsic value of
- 3:43its own. It's the only one we have. It's the
- 3:45only one in the universe. But conflict accelerates
- 3:50harm to the biodiversity, to wildlife, exactly
- 3:54when ecosystems are already under lots of pressure.
- 3:57And so for as long as this damage continues,
- 4:00there's a real need for organizations like SeaOps
- 4:04to exist. So SeaOps, the Conflicts and Environmental
- 4:08Observatory, we have a simple goal, and that
- 4:11is to reduce environmental damage from armed
- 4:14conflict and military activity. There are many
- 4:17ways to achieve that, but today I'm just going
- 4:19to focus on one of those, hopefully of most interest
- 4:22to the audience here, and that's monitoring and
- 4:25evidencing environmental harm in a robust and
- 4:28at times systematic way. Now, if you haven't
- 4:32come across us before, we are a small UK charity
- 4:35formed in 2018 and that was building on earlier
- 4:40work around toxic remnants of war and in part
- 4:42we emerged because of the big potential of open
- 4:46source investigation to give a voice to the environment.
- 4:512018 now maybe seems a bit of a bygone age given
- 4:54how dramatically the world around us has shifted.
- 4:58Geopolitical fragmentation, climate stress, AI,
- 5:01disordered discourse and also the growing securitisation
- 5:05of the environment. However, I think in that
- 5:08time the dial has shifted a bit. The environment
- 5:11is perhaps no longer the silent victim. as we
- 5:14used to say back in 2018, but we need to continue
- 5:17to work hard to ensure it's not a pyrrhic victory
- 5:21and that the environment does not now become
- 5:23the sidelined victim. So we need to sort of stick
- 5:26to our core values to keep on track. So that's
- 5:30solidarity, it's collaboration and it's expert
- 5:33situated knowledge. And yeah, I think at this
- 5:36point it'd be nice before you see all the pictures
- 5:38and images and stuff to give full credit to my
- 5:41colleagues. So we are comprised of environmental
- 5:44scientists, remote sensing specialists, policy
- 5:47analysts, legal experts, and based in the UK,
- 5:51the Ukraine and beyond. So for anyone in the
- 5:54audience thinking about Korea, maybe in investigative
- 5:57environmental work, it might seem very niche,
- 6:00but I think it's a growing field and there's
- 6:02lots of space to work in and lots of different
- 6:05paths. So what do we mean? What do I mean when
- 6:09I say environmental harm? and why is this so
- 6:12much different in war to in peacetime? Now, in
- 6:17conflict, environmental harm takes many forms.
- 6:19Some of this is really immediate and obvious,
- 6:22big explosions. Other forms are slower or more
- 6:26systemic, polluted soils, damaged water systems,
- 6:31or a collapse in environmental governance. But
- 6:34let's start with something very clear and obvious
- 6:36and direct, which is air pollution. Often what
- 6:40we see in almost every war is that oil facilities
- 6:44are targeted and this is just one example from
- 6:49earlier this year in Tehran where there's a nice
- 6:51satellite image here with two fuel storage sites
- 6:54that have been simultaneously attacked. This
- 6:57is one big difference to peacetime, you might
- 6:59have one of these incidents as a one -off. In
- 7:02war you get them simultaneously, very frequently.
- 7:07And whereas in peace time you have a big civil
- 7:09response to this, most of the time in conflict
- 7:14settings there is no civil response. There's
- 7:15no people, there's no bodies going out to measure
- 7:17air quality and provide PPE and safeguard health
- 7:20and so on. The air quality impacts, now maybe
- 7:24we don't know anything much about what's going
- 7:26on in the stratosphere, mass missile launches,
- 7:30you know, they could be depleting ozone. So there's
- 7:32different interesting research angles to go in.
- 7:34But maybe bringing things back down to the ground
- 7:37is exposure to air pollution that is very linked
- 7:42to war, so energetic compounds and repellents
- 7:46that you wouldn't, civilians would not be exposed
- 7:48to in peacetime. So this on screen here is an
- 7:52explosion of an ammunition storage site in a
- 7:54densely populated area. Sticking with this example,
- 7:58showing the damage from such an explosion in
- 8:00eastern Ukraine, you can also see how it may
- 8:03lead to widespread land contamination and energetic
- 8:07materials, for example, TNT. We know this is
- 8:10a possible carcinogen and exposure to even low
- 8:14concentrations can lead to long -term health
- 8:16impacts. You know, waste from these kind of explosions
- 8:20can also pose serious health threats. So even
- 8:23from the very building materials themselves.
- 8:26So these, for example, the Asbestos Mountains
- 8:28in Gaza. You could have imagined a child surviving
- 8:32everything that's gone on there, only to grow
- 8:34up and die from asbestosis. Examples of complex
- 8:39land contamination in urban areas, especially
- 8:42when different types of facilities with different
- 8:45pollution profiles are hit. So here's the area
- 8:48of about a kilometre by a kilometre in Bari Industrial,
- 8:51this area in Khartoum, which has destroyed plastic,
- 8:55pharmaceutical and food factories. Next to destroyed
- 8:59warehouses and car workshops, together this creates
- 9:02a very complex pollution mix. So land contamination,
- 9:07it's not just an urban problem. When the fighting
- 9:09is in fields it becomes a big problem for crops,
- 9:12both now and in the future. That speaks to food
- 9:15security, it speaks to livelihoods. Soil is physically,
- 9:19biologically and chemically damaged. Explosives
- 9:22like RDX can bioaccumulate in crops and pose
- 9:26reverberating threats for many many years. On
- 9:31top of this is the risk from unexploded ordnance
- 9:34and other military litter. So particularly at
- 9:38the moment there's concern from the impact of
- 9:41fibre optic drones and the impact they will have
- 9:43on wildlife in Ukraine especially. But there's
- 9:46also reports of them being used elsewhere now
- 9:48so we've seen this major change in material being
- 9:51used in war. which could have really significant
- 9:54long -term effects on wildlife. Even entire nature
- 9:57reserves have been transformed delicate ecosystems
- 10:01into military training grounds. This has happened
- 10:04a lot in Ukraine. So then we come to the very
- 10:09profound impact war has on water and its reverberating
- 10:13impacts. Just one example would be the militarisation
- 10:17and draining of the Mesopotamian marshes in Iraq.
- 10:21starting back in the 80s, not only altering the
- 10:25entire hydrological system, but also an entire
- 10:28culture. Nowadays we're seeing this over -resurgence
- 10:32in using water as a military tactic. Huge events
- 10:36like the Kokovka Dam collapse, but in addition
- 10:38to those very large events, there's also thousands
- 10:42of small -scale damages to water and sanitation
- 10:45infrastructure, and this can release pollution
- 10:48into both surface waters. but also down into
- 10:52groundwater where the impacts are stored for
- 10:55many years. Out of this thing called climate
- 10:58change and yes, wars and military activities
- 11:01contribute to this. We are really only just finding
- 11:05out the magnitude of these conflicts and military
- 11:08missions and more monitoring is urgently required
- 11:12and a big space to expand into for us into work,
- 11:16I think. But really I could have taken you through
- 11:18dozens of examples of environmental harm and
- 11:22gone into way more depth for the more indirect
- 11:24and reverberating impacts. But I hope those slides
- 11:28gave you a brief taste of the range of harm that
- 11:30we see. One overarching feature of environmental
- 11:36harm in warfare is the convergence of vulnerabilities.
- 11:41So damage, stacks upon damage. This then compounds
- 11:44everything else. the human rights risks, the
- 11:48displacement, the food insecurity, public health
- 11:51crises. And this all unfolds in places where
- 11:56data is scarce, governance is weak, and accountability
- 12:00can be contested. And this is exactly where OSINT
- 12:04methods come into their own. OSINT satellites,
- 12:09social media posts, eyewitness footage, and public
- 12:13dead sets. They give us a way into understanding
- 12:15how on the ground, I'm fairly sure everyone in
- 12:18this room and listening understands that. Now
- 12:22earlier in COB's work we carried out a lot of
- 12:24ad hoc investigations. This could be verifying
- 12:27an oil spill clean up, could be assessing ammunition
- 12:30storage explosions, flooding events, so on. These
- 12:34snapshots were valuable, but they were just at
- 12:38snapshots. They told us what happened that day
- 12:41to that site, or maybe over a few days. However,
- 12:45as we've just found out, conflict -related environmental
- 12:48harm is anything but a series of isolated events.
- 12:52It's dynamic, it's cumulative, and it's all interconnected.
- 13:00So this is why we've moved towards more systematic
- 13:02monitoring. OSINT remains absolutely central
- 13:06to this, and it's what led us to develop the
- 13:09approach I'll describe next. I'll bring you on
- 13:12to Wyzen. We'll soon be launching this formally,
- 13:16but today I'm going to give you a sneak peek
- 13:19of something we've spent the last few years developing.
- 13:22Wisen, the wartime incidents to environment database.
- 13:27The idea is straightforward. Conflicts generate
- 13:30countless environmentally harmful incidents,
- 13:33but the evidence has been scattered, inconsistent
- 13:37or disappearing. Wisen brings that information
- 13:41together in a structured, verifiable remote analysis
- 13:45framework, and assesses each incident's environmental
- 13:49risk and harm from the local level to the national
- 13:52scale. So I'd like you to think of it as a sort
- 13:56of counterpart to what organizations like Air
- 13:59Wars and ACLID have done in their respective
- 14:01domains. I think that's an outstanding idea,
- 14:05but it emerged in its earliest form in the early
- 14:07days of the full scale of invasion of Ukraine.
- 14:10when it was immediately obvious that the scale
- 14:13of environmental damage required systematic tracking.
- 14:16We had the tools, satellites, OSINT, environmental
- 14:19expertise, and why the Ukraine grew from there.
- 14:24Now, there are really many other important ways
- 14:27of studying conflicts in the environment, like
- 14:30mapping burned areas, vegetation loss, crater
- 14:32detections, air quality time series, all of these
- 14:35things, very important, we call them continuous
- 14:38cohort approaches. They're essential and academics,
- 14:42especially now, are doing fantastic work in that
- 14:44space. But WISON is the counterpart to that.
- 14:48It focuses specifically on discrete incidents
- 14:52affecting specific facilities. It's built around
- 14:57hierarchy. So at level zero, we have damaged
- 15:01facilities, but where we don't have enough temporal
- 15:03detail to define an incident. Level one is a
- 15:07wide ranging list of verified incidents. with
- 15:10calculated environmental risk scores. And level
- 15:14two is the big deep dive investigations into
- 15:17the highest risk incidents where we assess the
- 15:19actual environmental harm. When we cast a wide
- 15:26net, we do systematic searches of social and
- 15:29mass media, emergency services updates, curated
- 15:32data sets, OSINT community outputs, and satellite
- 15:36driven findings. to generate a list of potential
- 15:39incidents. We then verify those, geo and chrono
- 15:43locate them. And if it meets our definition,
- 15:46we give it a score for the environmental risk.
- 15:49And that's calculated based on the theoretical
- 15:54risk. So what the type of facility is and where
- 15:57it's situated. Is it near people? Is it near
- 15:59crops? Is it near ecologically important areas?
- 16:02And also an event magnitude score based on the
- 16:05the magnitude of what would have happened. We
- 16:11can see that these occur across Ukraine and Norway
- 16:16really escapes. This gives you a real sense of
- 16:20the breadth of incidents and the highest risk
- 16:23incidents we then promote up to level 2 analysis.
- 16:30Level 2 is where the serious OSINT work happens.
- 16:33For each priority incident, we build a structured
- 16:36dossier. So we have a chronology of social and
- 16:39mass media posts, capturing the earliest posts,
- 16:44multiple viewpoints, testimony, screenshots,
- 16:47URLs, all of that business. So we have a reproducible
- 16:52search trail. We also generate a stack of annotated
- 16:57satellite imagery before, during, after, from
- 17:01all of the openly available satellite feed sources
- 17:04and occasionally some commercial ones too. We
- 17:09then do more detailed geolocation of key footage
- 17:14even if we know it is definitely where it is
- 17:18it's good practice to do so and maybe a more
- 17:21unique element is we collect subsequent evidence
- 17:24from later imagery a month later two months later
- 17:28or from local reporting or from official assessments.
- 17:31And this means we can do an assessment of not
- 17:35only the immediate damage, but also the longer
- 17:36term damage. And that's exactly what we do. We
- 17:41have a sort of structured environmental harm
- 17:43assessment, looking at hazards, affected media,
- 17:47loss of ecosystem services and local vulnerabilities.
- 17:50So this gives us then a scorecard record of harm.
- 17:56And if the same facility struck multiple times,
- 17:59we can add that together. and work out cumulatively
- 18:01the most impacted facilities. The result is a
- 18:06detailed, scalable and transparent picture of
- 18:10environmental harm. We can see national patterns,
- 18:14we can see regional trends, individual facility
- 18:19histories. WISEN has already been used in our
- 18:24advocacy, media work and external investigations.
- 18:27and also for ground proofing with mine action
- 18:30teams. And because the methodology is transferable,
- 18:33we've also applied it in Khartoum and Al Jazeera
- 18:36states in Sudan. Our in now is to partner up
- 18:41so we can scale up, develop it more and roll
- 18:45it out to other appropriate armed conflicts,
- 18:49employing local situated expertise and ultimately
- 18:54and eventually handing responsibility. over to
- 18:57those local partners. Now, Wisen isn't perfect.
- 19:02No data set is, but it gives us something we've
- 19:04never had before, which is a replicable way of
- 19:06understanding environmental harm of scale in
- 19:09conflicts. And so this sets up the question.
- 19:13Now that we have this data, what can it help
- 19:15us change? So one is accountability. Once harm
- 19:21is documented clearly and consistently, we can
- 19:24use it for justice. This can be very simple in
- 19:28terms of recognition or for transitional justice.
- 19:31It could even extend as high up as international
- 19:34legal processes. What it does is it moves the
- 19:39conversation from we think this happens to here
- 19:41is the evidence and here is the scale. It can
- 19:45also support behavior change and norm development.
- 19:52So we can work out all the time which actions
- 19:55are causing the greatest environmental harm,
- 19:57which tactics, and then we can make the argument,
- 20:01maybe some of these tactics are not appropriate.
- 20:06And then finally there's recovery. If we know
- 20:07where contamination and ecological damage are
- 20:10concentrated, we can prioritise those areas for
- 20:12remediation, for restoration and for resource
- 20:16and funding basically. And this is central to
- 20:19any green and just recovery process. And that's
- 20:23something we're working on. actively in Ukraine
- 20:25at the moment. So to close I'd like to turn the
- 20:31question back to the audience actually. What
- 20:34are we missing? Are there other places OSINT
- 20:37can help? WISEM captures what we can observe
- 20:41but what are we not seeing? Could we be looking
- 20:44more at health impacts and assessments? More
- 20:48at prevention and predictive analytics? More
- 20:52at supply chains, the militaries for example
- 20:57and all of this in the context of the information
- 21:01space which is shrinking is becoming more fragmented,
- 21:04more content is ephemeral and there's more this
- 21:08and misinformation and that is they also impact
- 21:11environmental narratives. So how do we go about
- 21:15keeping documentation trustworthy? So genuinely
- 21:19welcome your ideas, your questions. And thank
- 21:22you for being here today. Thank you. Thanks for
- 21:27being patient as well with my clicking. Apologies
- 21:31about the tech issue there that prevented you
- 21:34from clicking yourself, but I really appreciate
- 21:36you taking us through that and actually taking
- 21:39us through some quite surprising to me anyway,
- 21:42impacts of war. Some of the things like air pollution,
- 21:45I'd not even thought about when we're covering
- 21:48conflict. Usually we look at fires. and deforestation
- 21:53as environmental impacts, but not necessarily
- 21:55air pollution, which was super interesting. We
- 21:57have a bunch of questions in the chat, but just
- 21:59to respond initially to your question at the
- 22:02end there, a lot of people are saying, what are
- 22:04we missing? Somebody said my answer, volunteers,
- 22:07and they are ready to volunteer. I don't know
- 22:10if you have a volunteer program or a way that
- 22:13people can get involved. It's almost like a set
- 22:17up question. So we previously have trialed volunteers
- 22:22like maybe like a couple of years ago and what
- 22:26we found was that we were probably asking too
- 22:28much of them in terms of like the there's the
- 22:32awesome knowledge but then there's like an extra
- 22:34bit of environmental like knowledge that's missing.
- 22:38So we've been sort of learned from that and we've
- 22:40been thinking about it and I certainly do think
- 22:44there is a role for volunteers and At the moment,
- 22:48we're trialing out with some sort of, some students
- 22:53are working part time on the level zero part
- 22:56of the database, which is just damaged facilities.
- 23:00So we're going to see how that goes and maybe
- 23:02see if that is something that we can roll out
- 23:05as a, as a wider thing for volunteers as a citizen
- 23:09science thing perhaps. Could it be hosted somewhere
- 23:12like Zooniverse? We're just, we're sort of, yeah.
- 23:15investigating these ideas, it would be good to
- 23:17know what's too much to ask of volunteers and
- 23:21what's the reasonable ask. That's maybe where
- 23:23we struggled before. Yeah, perhaps people can
- 23:28put their experience of volunteering in the chat.
- 23:31I know, for example, we have a volunteer group
- 23:33here at Bellingcat and it's always a learning
- 23:37process in terms of how much to ask for and how
- 23:39much to get back. What I often find is that people
- 23:42are very willing to give a lot of time and actually
- 23:45it's up to you as the responsible organization
- 23:49to make sure that they're not overworking, which
- 23:51I think is a key point. We've got a few questions
- 23:56about the Wison map as you presented it. Saiba
- 24:01has asked, that Wison map of Ukraine, does that
- 24:04show overall environmental damage or is there
- 24:07a way or reason? to separate it into air, water,
- 24:10ground and damage, et cetera. And somebody else
- 24:15has put, is there a correlation made between
- 24:17the WISEN findings and population health after
- 24:20an incident? For example, degradation in crops
- 24:23slash animal health, et cetera. So have you thought
- 24:26about breaking down the things mapped on the
- 24:30WISEN maps to show kind of the breakdown in different
- 24:33types of environmental damage? And also is it
- 24:36linked to population health as well? Absolutely
- 24:40fantastic questions. And yes, we can break down
- 24:43that map. We could separate it out by the type
- 24:46of facility. So if we just wanted to look at
- 24:48water facilities, we could have the same map,
- 24:50just looking at water facilities, same for agriculture,
- 24:54industry, military even. And we could start to
- 24:58see how things look different for different sectors.
- 25:02It's important to say that wouldn't give you
- 25:04the whole story for like water in Ukraine. If
- 25:06you were just looking at our our database, it
- 25:09would give you a lot of information and a lot
- 25:11of the detail. We also need to then be still
- 25:15looking at some of the more, I can mention the
- 25:18sort of like continuous cohort approaches. So
- 25:21looking at not incidents, but let's say water
- 25:24quality over a very long period of time in lots
- 25:27of different places, which is not something that
- 25:29would be picked up by Vyzen as an incident. You
- 25:32might see. the three, four years of the conflict,
- 25:36the change in water quality, which you could
- 25:38maybe attribute to the conflicts, but it's sort
- 25:40of one of these more hidden sort of things. Absolutely
- 25:44breaking it down by sector, by domain is really
- 25:47important. We're trying to understand this in
- 25:51terms of health. That's the sort of next big
- 25:56frontier. That's where a present is. exceptionally
- 26:02hard to make the connection between environmental
- 26:06damage, some exposure to pollution and the health
- 26:10impact. The sort of, even in peacetime in civilian
- 26:16settings, it's very hard to do that sort of epidemiological
- 26:18work because there's so many other confounding
- 26:22things happening. That's made even harder in
- 26:25conflicts because people's bodies are under more
- 26:29stress from the mental stress. physical stress,
- 26:32there's everything else going on. Um, but it's
- 26:38definitely, I think that where if we keep pushing
- 26:42and keep trying, we will like uncover lots of,
- 26:46lots of connections, lots of important connections
- 26:48and connections that are not just acute and over
- 26:55a few days or weeks, but like chronic over a
- 26:58very long term. More just in the way What's with
- 27:01pageant orange and herbicide use in Southeast
- 27:04Asia? You mentioned a lot of the time in your
- 27:08talk about the cumulative effects that you're
- 27:12monitoring. Where does your organization draw
- 27:15a line on monitoring? Are you monitoring these
- 27:19effects of conflict forever? When does it no
- 27:22longer become? When does conflict no longer become
- 27:26the causation and how long into the future are
- 27:29you monitoring this land and these pollution
- 27:33effects? I think that's an amazing question.
- 27:39And it depends what you're particularly interested
- 27:41in, but the answer can be decades. Even now,
- 27:46there's fields that farmers can't use in Northern
- 27:49France because of contamination from World War
- 27:51I battlefields. So you do need to think on those
- 27:55time horizons. And I also think there's a lot
- 28:00of space to research into some of the wars in
- 28:03the 60s, 70s, 80s, 90s, even into the 2000s of
- 28:10health impacts now. And there's been like the
- 28:12most work done around like US veterans who were
- 28:16exposed to burn pits in Iraq, Afghanistan and
- 28:20so on. where all of the outside the bases they
- 28:24were just burning everything and they now have
- 28:28elevated health risk let's say. What will happen
- 28:32to the local civilian populations in those places?
- 28:36We don't really know anything about that so we'd
- 28:39need more monitoring and more monitoring into
- 28:41the future. Where to draw the line? The line
- 28:46gets drawn for us I think by the resource we
- 28:50have. More hands on deck are needed by the sounds
- 28:54of it. And I'm sure everyone listening is now
- 28:58really fascinated and interested in getting involved
- 29:01in this kind of work. One of my colleagues, Jake
- 29:05Godin, who has been working at looking at the
- 29:09land changes in Gaza for quite some time now
- 29:12is curious if you or your colleagues have tried
- 29:15messing around with hyperspectral imagery yet
- 29:17on observing environmental changes slash impact.
- 29:20Perhaps you can give us an insight into some
- 29:22of the tools you mentioned, satellite imagery,
- 29:25but specific tools that you use on a regular
- 29:27basis to monitor this work. Yes. Okay. Let's
- 29:31start with hyperspectral because it's the most
- 29:33exciting. Um, it is, I think if we had access
- 29:40to hyperspectral data, so that's where you could
- 29:45have sort of the spectral fingerprint for every
- 29:48object. on the surface of earth basically. That
- 29:52would be really, really sort of transformative
- 29:56in environmental monitoring conflicts. Environmental
- 29:59monitoring overall, but like especially in conflicts
- 30:02we could much more easily document change. We
- 30:07haven't actually played with hyperspectral, at
- 30:11least high resolution hyperspectral, high spatial
- 30:15resolution hyperspectral. that we would love
- 30:18to if there's any commercial companies out there
- 30:21who would like to take us on board to do case
- 30:24studies with. What we are doing, otherwise, is
- 30:28pretty much the same as lots of people here,
- 30:30which is, you know, using the openly available
- 30:34stuff. So Sentinel -2, obviously, also Sentinel
- 30:40-1, Radar, Danger Detection, Muse. If we have
- 30:45to we use proprietary imagery so especially the
- 30:49planet scope stuff is good or we can buy some
- 30:54very high res imagery which we find we often
- 30:57need to do if we want to know if a certain like
- 31:01chemical tank has been damaged. You just simply
- 31:04can't see that from a Sentinel -2 but you might
- 31:06be able to from a very high resolution image.
- 31:11I expect a lot of you are familiar with that
- 31:13sort of stuff in the room. Maybe what you're
- 31:15not so familiar with is all of the sort of open
- 31:19environmental data, which is out there. Now this
- 31:22will sort of vary between countries, but there's
- 31:24a lot of derived products, maybe from satellite
- 31:27data, maybe from models. We see loads of value
- 31:34of combining the OSINT stuff with that open environmental
- 31:38data. And that's what helps us sort of be able
- 31:41to, to assess environmental health. I wonder
- 31:45if any of our listeners, and I'm speaking more
- 31:50directly to the Bellingcat staff members who
- 31:52are lurking in here, if they have any suggestions
- 31:55for satellite image tools or tools in general
- 31:59that might be useful to this group. I've got
- 32:02another question. Could you say more about the
- 32:05verification process? Do your local partners
- 32:07verify the incident and impact assessment? How,
- 32:11how do your local partners get involved with
- 32:13the verification process? So the way we've, the
- 32:17way we've done that so far is have a, uh, bring
- 32:24in as a staff member working on the database
- 32:27with situated knowledge and, uh, contact with
- 32:36people who, who are there, even if they're not,
- 32:39they might be diaspora. but they have contacts
- 32:41for people who are there so they can have the
- 32:45situator knowledge. They can tell us that's not
- 32:48a, that's not a farm, it's a military facility.
- 32:51But as to our eyes on satellite, it might just
- 32:53look like a farm, things like this. Um, it's
- 32:59something we need to be careful with, of course,
- 33:02from the ethics side, exposing people to things.
- 33:07Um, but generally that's how we get the the expert
- 33:10knowledge in for the situated knowledge and then
- 33:13for the rest of the verification it's always
- 33:15the usual triage. So making sure that there is
- 33:22alignment between different satellite products,
- 33:25the satellite products and the social media.
- 33:28And yet having quite high bar for saying something
- 33:33that's happened. You talked about justice and
- 33:38accountability as a next step. This might be
- 33:41a silly question, but who is responsible for
- 33:45protecting the environment during conflict? Are
- 33:49there international laws that you can appeal
- 33:53against and steps that you could take with evidence
- 33:58that you've collected to try and change things
- 34:01and make sure the environment is top? of the
- 34:04conversation when conflict happens. Ultimately
- 34:08the answer is basically no. There are provisions,
- 34:14you know, you could apply environmental damage
- 34:16as a Rome Statute war crime, but the thresholds
- 34:20which need to be met are severe, long -term and
- 34:25widespread. The definitions for those thresholds
- 34:29are not clear when it comes to the environment.
- 34:33And so yeah, no, there's been no successful case,
- 34:37environmental case. That might change. There's
- 34:41guidance coming out very soon from the ICC on
- 34:45better prosecution for environmental crime. So
- 34:49we'll see what they come up with. There's also
- 34:51the movement towards having ecocide as an international
- 35:00crime. which is something that may, may eventually
- 35:04come, but at the present moment and for what's
- 35:07recently happened in Sudan, Gaza, Ukraine, there's,
- 35:11there's nothing, but it's possible that will
- 35:15change, it's possible Ukraine will bring cases
- 35:17and there will be processes and the lawyers will
- 35:21lawyer and, and there will be sort of, yeah,
- 35:25maybe reparations, some sort of high level accountability.
- 35:31Maybe you can tell I'm not a lawyer. Well, I'm
- 35:34hoping that your organization is already speaking
- 35:37to legal institutions, but if anyone's listening
- 35:40and thinks that they might be the right fit,
- 35:42perhaps they can reach out to you. I think it's
- 35:45great though that you're collecting this data
- 35:47in preparation for such a step as well. Speaking
- 35:51of Ukraine, I just put some Bellingcat resources
- 35:55in the chat. Back in 2023, we built a tool that
- 35:59shows what war has done to Ukraine's forests.
- 36:02It's a deforestation tracker tool that you can
- 36:05use. And more recently, we've just had a guide
- 36:09written on visualising impact on the environment
- 36:14from conflict, including displacement data that's
- 36:17linking to the question about population health
- 36:20as well. We've actually had a question on the
- 36:23fibre optics waste that you mentioned briefly
- 36:25at the beginning. Someone's asked, can you talk
- 36:28a little bit more about the impact of fibre optics
- 36:31waste on the environment from the Ukrainian conflict.
- 36:33I think this is quite a new concept for quite
- 36:35a few people in the audience. Yeah. So I'm guessing
- 36:42people have seen footage of these fibre optic
- 36:45drones. They have a little spool and that may
- 36:47have 10 to 40 kilometres worth of fibre optic
- 36:51cable on it. And there's thousands of these,
- 36:55well tens of thousands of these drones flying
- 36:57around the front lines of Ukraine. They use fibre
- 37:00optics so that they can't get jammed from electro
- 37:05magnetic interference and what we then see is
- 37:10this, these fibre optic cables grouped across
- 37:14trees, upland. It's really strange when you watch
- 37:19a lot of the footage from the conflict they kind
- 37:21of glisten and they look almost quite pretty,
- 37:25like quite Christmassy, but I think what we're
- 37:27going to find is that they have a sort of similar
- 37:29impact on wildlife, like, um, so fishing nets,
- 37:33um, for fisheries and so on, because they're,
- 37:38they end up in a tangle. Wildlife will get trapped,
- 37:43but wildlife were also using these. So there's
- 37:45like some images of like a bird's nest made out
- 37:49of fibre optic cables rather than twigs. pretty
- 37:54devastating. Um, and yeah, these are, these are
- 37:59all draped over areas of frontline, which are
- 38:02heavily mined, inaccessible, and they're going
- 38:05to be there for, you know, maybe decades. They're
- 38:08not going to degrade very naturally, very quickly.
- 38:11It's going to be a very long -term problem I
- 38:12think. It actually brings me to my next question,
- 38:17which is what do you think is a long -term implication
- 38:22of conflict -related environmental degradation,
- 38:26which is most underestimated. The thing that
- 38:29you are concerned is going to be unraveled in
- 38:33a couple of years time as a major issue. That's
- 38:39a really good question. I do not have an immediate
- 38:40answer for. I think that I don't know. I'm going
- 38:51to give an example. I'm not going to say this
- 38:52is the most, um, significant, it's just one that's
- 38:55immediately come to mind. So I'm thinking in,
- 38:59in Ukraine, we have Krakow -Kadan collapse. We
- 39:02have the big flood. We also have the, the upstream
- 39:06area, which relied on the reservoir for all the
- 39:09irrigation. And that's now gone. There's severe
- 39:12drought. There's going to be probably, even if
- 39:17there is access in the future for farmers, civilians
- 39:21to return. It's going to be really hard to make
- 39:24a living there. There's going to be lots of soil
- 39:25erosion because of climate change. So actually
- 39:28southern Ukraine is going to be sort of very
- 39:32severe impacts from climate change. And sort
- 39:36of there's a risk that areas like that will be
- 39:41because of the environmental impacts of war become
- 39:44very, very hard to live and survive in. that
- 39:49might be a very long -term consequence, like
- 39:52people move permanently. And I'm guessing that
- 39:57that impact isn't just isolated to Ukraine either.
- 40:01There's going to be other regions similarly affected
- 40:03in the same way. It's very, very difficult to
- 40:07think about positives. And I just wanted to remind
- 40:09people that there's about 10 minutes left. So
- 40:11if you want to put your comments in, your last
- 40:13final questions, please do. And thank you to
- 40:17those who've been dropping links in the chat
- 40:19as well. I'll make sure that they're included
- 40:21in the description of the podcast. So if you
- 40:23do have any helpful links related to this field
- 40:25of work, please do pop them there. I was talking
- 40:29about trying to pull some positives from this.
- 40:31What gives you hope in this field? If that's
- 40:34an impossible question, please respond as such.
- 40:39But is there any hope within this field that
- 40:42you've seen from your research? Yes, yeah, there's
- 40:48got to be hope. There's always hope. We're here
- 40:51today talking about this and that gives us hope,
- 40:55right? There's like people who are affected directly
- 41:01by environmental damage in conflict. They are
- 41:06a source of hope because they do things on the
- 41:09ground to make their environment better. What
- 41:13we need to do is make sure they get the support.
- 41:16to do that in the right way, in the safe way,
- 41:19in a sustainable way. And I think that is possible.
- 41:26That might come. There are changes going on at
- 41:28a high level as well. You know, things like the
- 41:31move towards the eco side. These will help. Yeah,
- 41:37we always have to have help. It keeps us driving
- 41:42forward at least at Bellingcat. As we're talking
- 41:47about that, you've mentioned within your talk
- 41:49that OSINT is kind of a hopeful method that's
- 41:52been helping advance this. What would be your
- 41:58advice to people who are listening to this, perhaps
- 42:00with open source research skill sets like satellite
- 42:06analysts and people who are very used to kind
- 42:09of looking at environmental degradation through
- 42:12open source methods. What would be your advice
- 42:15to them in terms of getting involved in the kind
- 42:18of work that you do, not necessarily related
- 42:20to your organization, but within the same field?
- 42:24What would be your advice to them? Any tips or
- 42:26tricks or organizations that you might want to
- 42:29point out that people could reach out to as well?
- 42:42try and learn as much as you can about the science
- 42:46of fires, the science of deforestation, ecology,
- 42:51like the more knowledge you have on a particular
- 42:55environmental topic, the better your analysis
- 42:58is going to be. So I think don't shy away from
- 43:02learning stuff as well. Don't just think that
- 43:06because you You can write a Google Earth engine
- 43:09script that detects burn areas you know about
- 43:12fires. Like go away, teach yourself, learn, there's
- 43:16stuff out there. Get in contact with organisations
- 43:19like us, like we're always happy to answer an
- 43:21email. So please do. And then give it a try,
- 43:26yeah. See what you can do. Publish it. There's
- 43:30lots of, there's like, I can't like reiterate
- 43:34enough how much is not being reported and there
- 43:38is to report upon. So yeah. Where do you see
- 43:46the biggest data gaps? Like if you were, if you
- 43:49were wishing on the spot right now for more researchers
- 43:52to dive into a certain area of what you've spoken
- 43:55about, where are the huge data gaps within this
- 43:59field? So I can point to like countries where
- 44:04there's massive gaps of information. So Sudan
- 44:07is an obvious one. It's probably an answer you've
- 44:10heard in relation to not just the environment,
- 44:13but there's other places too, places that are
- 44:16maybe not so actively at war that just come out
- 44:18with fighting, places like Yemen, Syria, Lebanon,
- 44:23where there is so much that we could know and
- 44:27need to know, but you're not currently know and
- 44:30people are not researching. If I had to pick
- 44:34an environmental domain where we don't know very
- 44:38much, I would probably go with biodiversity.
- 44:45And not on vegetation, not habitats, but more
- 44:47on actual wildlife itself. How particular species
- 44:52are affected by something like noise pollution
- 44:55from water. We don't know anything about that.
- 45:01Really. You mentioned Sudan. I've put the Sudanese
- 45:06Civil War link in the chat there for anybody
- 45:10who wants to chat, well, research, particularly
- 45:14that particular conflict. Don't be afraid to
- 45:19also put environmental damage in these conflict
- 45:24channels that we have in the server. It's absolutely...
- 45:28It's absolutely something we want to talk about.
- 45:30It's absolutely fine to not focus so much on
- 45:33the immediate effects of war, but the long term
- 45:36effects of war. That's absolutely vital as well.
- 45:42Cyber's written, good grief. I forgot about noise
- 45:44pollution. Yeah. That's what I found through
- 45:47this talk is that more and more Owen, you're
- 45:50uncovering all these types of pollution that
- 45:52we may have overlooked over this time. Have you
- 45:56ever had the most surprising kind of effects
- 46:01of war? What would you say in terms of environmental
- 46:04damage has surprised you when it's come to conflict?
- 46:08Did you always expect to, for example, find fibre
- 46:11optic cables being polluted via conflict? Has
- 46:16there been some surprising aspects that you've
- 46:19come across in different conflicts around the
- 46:21world? It's a good one. That's a good one. I
- 46:30think the sort of conflict is maybe not directly
- 46:37conflict but what I maybe didn't have so much
- 46:41of an appreciation of before was to how much
- 46:46the oceans are like the Wild West and what happens
- 46:50in the oceans or dodgy dealings there are. We
- 46:56know a bit more about this now, like all the
- 46:57great work about ship to ship transfers and so
- 47:00on. I think that was quite surprising to me.
- 47:05Learning about the amount of marine pollution
- 47:09basically. And linked to that is old marine pollution
- 47:16as well. So the amount of dumped unexploded ordnance
- 47:21from World War II in the Baltic. in the Irish
- 47:27Sea and how that's now degrading, corroding with
- 47:34warmer oceans and is potentially going to be
- 47:36a problem as things that are encased, TNT, et
- 47:41cetera, are released into the environment. So
- 47:45yeah, maybe, yeah, I'll go with marine pollution.
- 47:49Yeah, if you're interested in looking more into
- 47:53maritime issues and ship -to -ship transfer,
- 47:56I mean, we've recently had a fantastic stage
- 47:59talk from Christine Panton, who's just an ocean
- 48:03hobbyist, has a full -time day job, but in his
- 48:07evenings creates a really amazing tool called
- 48:10Sketchy Boats, which looks at suspicious boats
- 48:13doing suspicious things around the world. And
- 48:16we've also, Bellingcat created a tool, well,
- 48:20one of our contributors, Ollie Ballinger, who's
- 48:23a fantastic lecturer in London, he created a
- 48:26tool that's called the Ship Detection Tool, which
- 48:29also highlights using synthetic aperture radar
- 48:34imagery when ships are in port or doing ship
- 48:38-to -ship transfers in the middle of the sea.
- 48:40So you can have a look at that as well. I put
- 48:42the link to that in our chat and I'm sure SettleKnife
- 48:46or any of the mods listening could also put the
- 48:48link to the last stage talk on suspicious boats
- 48:52in the chat as well. We are coming close to time.
- 48:57I just want to read a final few comments. Someone's
- 49:02put, what we are doing right now is actually
- 49:04a step in the right direction. Spreading awareness
- 49:06is a very real action you can take. By keeping
- 49:09yourself informed, you can inform others. Yes,
- 49:12hopefully that's exactly what this stage talk
- 49:15has done for you guys listening in. And somebody
- 49:21else put, more on climate change, the amount
- 49:23of CO2 released from burning refineries and ammunition
- 49:26depots must be astounding, as well as local environmental
- 49:29effects and petroleum chemicals as well. Yeah,
- 49:32I think that was mentioned also within the talk.
- 49:34Right. Thank you so much, Owen, for your time
- 49:37today. I've really appreciated you coming in
- 49:40and walking us through what is practically quite
- 49:45a depressing view of the impact of conflict on
- 49:50the environment, but a very important one and
- 49:52the work that your organization is doing is incredible.
- 49:58Just to finish off, how can people get in touch
- 50:00with you? Where is the best place to find information
- 50:04on the observatory. How can people reach out
- 50:08if they want to help or offer tools for your
- 50:11work? and yeah, happy to speak to people. Great.
- 50:40If you want to continue this conversation, I
- 50:42mentioned the Sudan channel earlier up, but we
- 50:45also have a hashtag environment and wildlife
- 50:49as well as many conflict channels within the
- 50:53space. Please do jump in. To those, if you want
- 50:57to continue this conversation, highlight tools
- 50:59and resources that you can use to continue investigating
- 51:02this topic. But for now, thank you so much, Owen.
- 51:05And we'll be back in two weeks time with another
- 51:08stage talk. Thank you, everybody. Thank you for
- 51:14listening to the stage talk. If you'd like to
- 51:17catch a stage talk live where you can ask the
- 51:20guest questions, join the Bellingcat Discord
- 51:23server by visiting www .discord .gg slash Bellingcat.
- 51:30The music you've heard is titled Dawn by Newer
- 51:33Self and is courtesy of Artlist.