Intern Rachel Pierce reports on her work at SkyTruth.
Prospecting
I first heard of SkyTruth at least a decade ago. I was amazed! Like the Eye of Sauron (but for doing good), they monitored polluters and assessed biodiversity. All right around the corner from my classes in that tiny place called Shepherdstown. I just knew: one day I had to get in. Well, my fellow earthlings, this eagle has landed. With a Remote Sensing course under my belt, among other graduate-level GIS coursework, I felt more driven than ever to be a part of this eclectic crew. The unique personalities and semi-casual, remote work atmosphere contrast sharply with the grim reality of why we do what we do.
We can’t do anything to return Appalachia to its original splendor, but we are doing everything we can to improve how we monitor the extent of surface mining as well as post-mining performance. I was excited to collaborate with SkyTruth’s Geospatial Engineer, Christian, and fellow intern Ethan, on how to approach automated detection methodologies.
As you may already know from our previous work, surface mining — and Mountain Top Mining (MTM) specifically — is the most prevalent method of coal extraction in Central Appalachia. It comes with its own history that’s as twisted as the local bedrock. Over the past two years we worked to examine the complex legacy of environmental implications of MTM, in collaboration with Appalachian Voices and which we explored further with our latest publication in Restoration Ecology. We also continue to publish annual updates to our MTM dataset to continue highlighting the ongoing impacts mining has on Appalachia.
As a part of our commitment to highlighting the impacts of MTM, we are developing new approaches to improve the overall performance and accuracy of our detection model. We also want to find out to what extent mines may operate outside of their permitted boundaries. Considering these issues, we are exploring the potential of how utilizing available permit data can improve our overall detections, while also highlighting areas of concern.
Digging Into the Issues
To address the issue of identifying areas of concern, we can frame the problem around the objective: To find out where coal mines may be overstepping designated permit boundaries and to evaluate the area of overstep compared to the accepted tolerance. This complex situation involves a combination of legal, environmental, and geographical assessments:
- What are considered allowable tolerances to mining outside of a permit boundary?
- Who are the mine operators and where are the suspected facilities operating out of tolerance?
- Are the relevant permit documents up to date?
- Do the state agencies update their spatial information regularly?
Reviewing permit documentation and utilizing spatial data in concert with satellite imagery is a practical and useful method for evaluating the extent of any overstep compared to accepted tolerance, as a similar methodology was implemented in the aforementioned post-mining reforestation assessment. Because we can filter active, inactive, or another mine status, we will continue to use coal surface mine permit boundaries to bolster what we hope will become a ML pixel classification training dataset.
False Positives are an issue because they depict areas that exhibit features easily mistaken for an active surface mine. Types of false positive detections include unreclaimed sites, such as where mining activity has recently ceased, artifacts that appear in imagery despite efforts to filter them out completely, and barren landscapes, like open rock face, eroded soils, or even seasonal changes in greenness.
Shortfalls in spatial data collection are every data hunter’s nightmare and are ever-present in our world, despite modern marvels such as the Internet of Things and Software as a Service geospatially linking us all. Completeness, accuracy, and even accessibility are still very real issues that limit the ability of analysis to have trustworthy and impactful results. This is one of the biggest hurdles in conservation work, often paired with time, funding, and ground-truthing constraints.
Alas, we persevere.
Finding out the accepted overstep tolerance for each state is not so easy either. Without any standard, quantifiable limits, any surface mine detections that occur outside permitted bounds will need to be manually scrutinized. The area of overstep can be totaled and we can look for potential trends or patterns in the data. Opening lines of communication with relevant organizations and agency representatives can help bridge this knowledge gap, provided the information is available and accurate.
Maintaining data integrity is important at every step, so it’s paramount that we preserve the original data. While permit data can be sourced from a single entity — such as through the Geomine application from the United States Office of Surface Mine and Reclamation — it is better to go directly to the source, meaning we run into a lack of data standards and continuity across the board. The reasoning is that each Source Agency will have the most up-to-date records, likely with regularly scheduled updates. For example, Virginia’s Department of Energy makes daily updates to their online datastores, and mine permit data is updated regularly throughout a mine’s life cycle (Daniel Kestner, Personal Communication, Dec 2023). The Geomine data hub happens to contain Virginia’s coal surface mine permit boundaries in their composite dataset; however, during our comparison of current permit bounds to prior detections, we found a considerable gap in this data.
To ensure our newly acquired data maintains its integrity, we must create a new ‘normalized’ dataset of all the permit boundaries within the study area. In doing so, we will be able to refer to the data should something come up during extracting and transforming the permit boundary datasets. This can easily be done by selecting relevant records and exporting them to a new file within a database or directory. This process with the normalization, is known as Extract, Transform, and Load (ETL). We select and export (extract), add new attribute fields and populate them (transform) and, finally, export the data into a repository (load).
But what does ‘normalized’ mean, you ask?
Chipping Away at the Goods
Each state has its own set of procedures and naming conventions when it comes to generating spatial data such as permit boundaries. As a mine’s life cycle is dynamic, there are likewise many attributes describing the mine, its status, its type, and so on. But because there are no standardized methods of attributing a mine or its permit boundary, we need to create a dataset with consistent attributes. This is called Normalizing the Data. This means we will look at all the separate permit data (each state’s permit boundary polygons) and create new attributes that describe the permits: State Name, Permit ID, Permit Status, Permittee, Mine Name, Mine Operator, and Source Agency. To populate these new attributes, we can fill in things like State Name and Source Agency, then we’ll use Structured Query Language (SQL) to write expressions that reference the raw data and fill in the corresponding blanks of our new attributes.
Each state’s department of mining delegates different naming conventions for permit statuses, and some states, such as WV, go into excruciating detail while others maintain simpler designations. Alternatively, Virginia maintains separate and distinct datasets on permits with released status, bond forfeitures, and bond statuses. All considered, simplifying permit status to ‘Active,” ‘Inactive,” and ‘Other’ will suffice. Along the way there are bound to be bottlenecks and roadblocks. For instance, the question arose regarding whether a mine’s status has the same technical and legal meaning as permit status. How does one approach this without losing valuable resources? Sometimes an approach is best taken while noting these considerations for future work and continuing as planned. Other times, it means starting over at square one. Careful assessment of each attribute and its definition is necessary early in the process here.
Once we have our new attribute columns in place, and relevant data entered or copied into the empty values, we can finally merge the cleaned, normalized data into a singular unit, where a user can filter permit boundaries by any of the new attributes and our new, beautiful, and clean dataset will remain shiny and intact.
The Haul
The hope is this cleaned permit dataset will feed into the work we aim to do in terms of assessing post-mining performance. If we’re to improve the accuracy of future detections, we must assess the false positives in those prior. Inactive mines will be classified as non-mine pixels. The detections will be passed through a vegetation index (the well-known NDVI algorithm) to ensure a lack of greenness as we expect zero vegetation in an active surface mine. By manually reviewing the prior detections for any of the false-positive scenarios discussed in the section above, we can further train a ML model to evaluate detections based on Permit Status (Active, Inactive, Other).
Stewardship
The effort put into cleaning this permit data, and that of a future ML model, will benefit ongoing work in assessing post-mining performance. By improving our detections while monitoring these mining activities, we continue to keep the spotlight on operations that overstep their bounds, raise awareness of the impacts and dangers of MTM, advocate for those most impacted by MTM, and educate the public on how MTM affects those of us downstream.
The work doesn’t stop here. Mining activity is staged in bonds and while the staging isn’t as important for this aspect of the work in improving our detections of active mining sites, bond release status will be important for reclamation assessment. As we move toward new approaches that leverage this new mine permit boundary dataset, it will be made publicly available.
Working with and learning from the incredibly talented, intelligent, diverse group of driven people at SkyTruth is an extraordinary experience, one I’ll carry with me onward and upward. I am proud to yell it from any mountain top, removed or not.
SkyTruth’s New Report Finds 750+ Oil Slicks in Mediterranean Sea Revealing Chronic Pollution Problem
/in General /by Kelly FranklinGreece, Home of ‘Our Ocean Conference’ Which Started Yesterday, Had Highest Number of Slicks in Its Exclusive Economic Zone (EEZ)
(Athens, Greece – April 15, 2024) – A new report released yesterday at the Our Ocean Conference in Greece by SkyTruth identified 757 oil slicks covering 1.9 million hectares in the Mediterranean Sea between July 2020 – January 2024. Most of these slicks were discharged from vessels in transit. The data comes from our new online platform, Cerulean, the first free, digital tool that uses AI and satellite imagery to track ocean oil pollution and its potential culprits.
The exclusive economic zone (EEZ) of Greece, where the Conference is taking place, had the highest number of slicks (182), followed by Italy (163) and Egypt (102).
“Greece is plagued by this pollution problem, but it’s hardly alone,” said John Amos, CEO of SkyTruth. “This is a global issue that will take international cooperation and vigorous, transparent action to address. Vessels dumping toxic oil into the ocean is not new, but our ability to detect this formerly hidden behavior with Cerulean now leaves polluters nowhere to hide. Raising visibility is the first step to solving the problem.”
Additional Key Findings from Today’s Report:
The Mediterranean Sea is a biodiversity hotspot, home to approximately 11% of all marine species in less than 1% of the global marine area. Furthermore, around 20% of these species exist only in the Mediterranean.This includes whales, dolphins, porpoises, loggerhead and green turtles, monk seals, and more than 80 species of sharks and rays. Despite its immense importance to the global ecosystem, the Sea is facing huge pollution-related risks and is already warming 20% faster than the global ocean average due to climate change. Over the last 50 years, marine mammal populations in the region have decreased by 41%.
MPAs were created to protect this unique nature hub, but there is still a long road to go before regulation efforts are sufficient to meet the goals enumerated in the 30×30 target. Adopted in 2022 by nearly 200 nations, the 30×30 agreement aims to protect 30% of the world’s ocean and land areas by 2030 (“30×30”).
In support of that effort and in addition to the Mediterranean analysis, SkyTruth also today released the “30×30 Progress Tracker,” a new digital platform bringing together key marine biodiversity data. With support from the Bloomberg Ocean Initiative, the Tracker is the first free, interactive platform designed for the general public to see — at a glance — how well the world is doing in enhancing ocean protection globally in line with the 30×30 goal. The platform can be used by civil society campaigns to accelerate country-by-country participation, and by government agencies and policymakers to gain insights on 30×30 and track regional and global progress. The goal is to make information about 30×30 more accessible and transparent to everyone, particularly people in communities impacted by this ambitious conservation initiative.
“At SkyTruth we aim to make the invisible, visible,” said Amos. “At Our Ocean we’re hoping to raise awareness about chronic oil pollution as well as efforts to protect marine biodiversity. The first step to progressing on these issues is for people to be aware.”
View the full report – Mediterranean Sea Chronic Oil Pollution Analysis: July 2020-January 2024.
New Free, User-Friendly Platform for Tracking Marine Biodiversity Protection is Now Available Online
/in General /by Kelly FranklinOpen-Source Tool from SkyTruth and Bloomberg Ocean Initiative Aims to Make Global “30×30” Conservation Goal More Accessible, Transparent, and Actionable
(April 15, 2024 – Athens, Greece) – Today at the Our Ocean Conference a new digital platform was unveiled, making it easier for anyone to track progress on protecting the ocean’s biodiversity.
Built by SkyTruth with support from the Bloomberg Ocean Initiative, the “30×30 Progress Tracker” is the first free, public, interactive platform designed for the general public to see — at a glance — how well the world is doing on enhancing ocean protection globally in line with the goal of protecting 30% of the ocean by 2030 (“30×30”). The platform can be used by civil society campaigns to track country-by-country progress, and by government agencies and policymakers to gain insights on 30×30 and compare countries’ progress. The goal is to make information about 30×30 more accessible and transparent to everyone, including those in local communities who are directly impacted by this ambitious conservation effort.
In December 2022, over 190 countries adopted the 30×30 target under the landmark Kunming-Montreal Global Biodiversity Framework, marking the biggest conservation commitment the world has ever seen: to conserve at least 30% of the Earth’s land and water by 2030. But until now it has been very difficult to track the progress or effectiveness of this initiative, especially for non-expert audiences.
“As an organization that specializes in the democratization of data for conservation, we created the 30×30 Progress Tracker to give government agencies, advocacy organizations, and people who are directly impacted by the creation of protected areas an easy way to understand and communicate progress toward 30×30,” said John Amos, CEO of SkyTruth. “We aim to offer data as the foundation for broadening public participation and dialog, ensuring bottom-up pressure on the world’s governments to stem the tide of extinction while respecting the rights and knowledge of local communities and Indigenous groups.”
The ocean is at the front line of the climate crisis. It protects the planet from the worst impacts of climate change by absorbing 90% of the excess heat caused by emissions from human activity. But as global heating intensifies, the ocean is changing faster than it ever has, and its ability to regulate global climate and weather patterns is weakening. Wildlife populations have declined by 69% globally in the last 50 years, and 90% of big ocean fish have been lost in the last century. Science is clear that protecting and conserving at least 30% of the planet is critical to safeguarding the vital ecosystems we depend on for our health and livelihoods. Yet today only 15% of land, 15% of freshwater, and 8% of the ocean are protected.
“Access to free, reliable, and real-time data is critical to ensuring effective marine conservation and fighting climate change,” said Melissa Wright, who leads the Bloomberg Ocean Initiative at Bloomberg Philanthropies. “We are proud to support SkyTruth’s 30×30 Progress Tracker. Now governments, civil society, and local communities can drive progress towards the global 30×30 target, expand marine protected areas, and safeguard the health of our ocean.”
The tracker is not a replacement for the bevy of existing 30×30 data providers (e.g., World Database on Protected Areas, Marine Protection Atlas, ProtectedSeas Navigator), but rather an independent aggregator that is partnering with these providers to drive increased attention to the full ecosystem of tools. The Tracker will add more data as it becomes available, including socio-economic information and Indigenous territories. While the marine tracker is available now, SkyTruth plans to launch the terrestrial component at the COP16 United Nations Biodiversity Conference in October 2024.
“In the face of the planet’s alarming biodiversity loss, protecting 30% of the ocean by 2030 is one of the great callings of our times,” said Peter Thomson, UN Secretary-General’s Special Envoy for the Ocean. “To succeed in this endeavor, we need the right tools. By bringing multiple sources of data together, the 30×30 Progress Tracker offers a new and powerful platform empowering practitioners and advocacy organizations around the world to identify and map contributions toward 30×30. I trust this democratization of data will not only help communicate conservation efforts but will also bring transparency and accountability to the task at hand.”
Meeting 30×30 requires governments to establish robust Marine Protected Areas (MPAs) or ‘other effective area-based conservation measures’ (OECMs). Last year’s adoption of a new High Seas Treaty provides a mechanism to ensure that protections can also be implemented in the 60% of the ocean that lies beyond national waters.
The new platform not only enables users to see the location of currently protected and conserved areas but also allows them to overlay the location of key habitats and areas recommended by expert analysis as critical places for protecting marine biodiversity.
For example, the Mediterranean Sea, where the Our Ocean Conference is taking place, is a world biodiversity hotspot and harbors around 11% of all marine species in less than 1% of the global marine area. However, the Sea faces numerous pollution threats and is warming 20% faster than the global ocean average due to climate change. The 30×30 Progress Tracker allows users to quickly see both how much of the Sea is reported as protected versus the amount that is independently verified; as well as what key habitats fall within the conserved area to better understand whether the correct areas are being protected, and if there are others that should be considered.
“2030 is around the corner,” said Amos. “We need all hands on deck to make this a success. The time is now.”
Other expert quotes on the 30×30 Progress Tracker:
Brian O’Donnell, Director, Campaign for Nature: “The 30×30 Tracker is a fantastic tool to help governments, advocates, and scientists measure progress towards the essential target of protecting at least 30% of the world’s land and ocean by 2030. It is important to know the current state of progress so that we remain on pace to safeguard biodiversity while we still have the chance. Campaign for Nature looks forward to using it in our future campaigning efforts.”
Rita El Zaghloul, Director of the High Ambition Coalition for Nature and People: “As we work together in the implementation of the protection and conservation of at least 30% of the planet’s land and 30% of the ocean by 2030, it is vital to have accessible resources, like the Skytruth 30×30 Progress Tracker, that provide the data to empower communities across the globe to make informed decisions to advance 30×30.”
Tony Long, CEO, Global Fishing Watch: “Transparency and technology play a vital role in achieving the ambitious goal of safeguarding 30 percent of the ocean by 2030. With only a handful of years remaining to reach this target, we must accelerate progress–and we must do it together. We have to gain a more comprehensive understanding of what has already been protected and what yet needs to be designated. SkyTruth’s 30×30 Progress Tracker will be a valuable complement to our marine manager portal, which provides managers and researchers with open access to scientific data and analytical tools crucial for monitoring and protecting marine ecosystems.”
Jacqueline Savitz, Chief Policy Officer, Oceana: “Strong marine protected areas can increase ocean abundance and biodiversity within and well-beyond their boundaries. The 30×30 Progress Tracker will allow governments, journalists, NGOs, and researchers to track the data, monitor progress, and maintain accountability as we strive to truly protect 30% of our ocean ecosystems by 2030.”
Visit 30×30.skytruth.org to learn more and get involved. Share any feedback on improvements that could be made. And please spread the word. The tools and technology to protect our planet will only be effective with widespread adoption. This one is free and easy.
Historic data provides the basis for new analysis of mining’s impact on Appalachia
/in Mountaintop Mining /by Christian ThomasSkyTruth is making large strides in understanding the ecological history of mountaintop mining. In the year to come, we plan to dive even deeper.
Areas of elevation gain (blue) and loss (red) at the Hobet mine site.
Those of you who’ve been following our work will recognize winter as the time of year that we often release updates to our Mountaintop Mining dataset, or give updates about the new directions that we’ve taken this work. We have continued to expand our mining program area to better understand the complex legacy coal plays in Appalachia, and are setting up for a year of even more robust analysis to come. As long as mining mars the rivers and forests of Appalachia, we will continue to monitor it and work to create data that allows communities, regulators, and researchers to hold mine operators accountable for their recovery obligations.
2022/2023 Update
As part of our annual update to our dataset, we found that in 2022 75 km2 of new land was impacted by surface mining. We’ve also made some slight changes to the MTM data that we provide, by publishing data through the most recent full year. Our methodology, described in our 2018 paper, employs a data cleaning step that uses data from preceding and subsequent years to remove errors in detection: if a pixel is classified as a mine in 1985, non-mine in 1986, and mine in 1987 the data for 1986 is corrected to classify it as a mine. The opposite is also true for pixels classified as non-mine > mine > non-mine. We are now providing the un-cleaned data from the most recent year, 2023, as provisional detections. While any new land clearing due to mining is dramatic, the relatively small area cleared in 2022 highlights the continued decline in coal production across the region. As this decline continues we are shifting the focus of our work to explore the long-term geographical and ecological impacts of MTM operations across Central Appalachia. As we shift our focus, we are still committed to providing regular updates to our mine footprint dataset. To accomplish this without taking unnecessary staff time to run and process these updates, we worked to create an automated processing pipeline that drastically reduces the amount of time needed to prep our updates.
Landsat MSS
In 2023 we also made significant strides in understanding the full impacts of surface mining through the integration of early Landsat data into our detection methodology. By incorporating imagery from the Multispectral Scanner (MSS) sensors carried on the first three Landsat satellites, we are now able to detect mining as early as 1973. Our methodology for detecting mining from MSS imagery mirrors our existing approach to mine detection in the region. However, working with data from the old MSS sensors onboard Landsat satellites 1, 2 and 3 presents unique challenges when compared to working with the more advanced Thematic Mapper (TM), Enhanced Thematic Mapper (ETM), and Operational Land Imager (OLI) sensors carried by the later Landsat satellites (4-9). Landsat MSS data is provided at 60m resolution, as opposed to 30m for TM/ETM/OLI sensors. These early satellites also have smaller data archives, so there is less imagery per year than there is in later years. These factors contribute to a lower overall accuracy for our detections from 1973-1984 than from our 1985-2023 detections, so we are releasing the new data from 1973-1984 in a provisional (“beta”) state. We have plans to improve the data in the future making use of improved access to mining permit data, as well as newly available historic elevation data for the region. As we provide this data in its beta format, we welcome any input users may have about ways to improve our detections or regions where this new data performs poorly.
Images above detail the challenges presented by Landsat MSS imagery compared to Landsat TM imagery. All data is from 1985. The top row displays TM data, which is higher resolution; and because there is more TM imagery available, mined lands are more distinct on multi-image composites. The bottom row shows a corresponding MSS data composite for comparison.
Elevation
In 2023 we also began exploring the topographic impacts of MTM. Through the use of an assortment of publicly available elevation datasets, we are now working to analyze the volumetric effects that mountaintop removal mining and valley fills have on the Appalachian landscape. Working with elevation data provides its own set of challenges to overcome in order to conduct helpful analyses. These data lack the temporal resolution of imagery: while the Landsat satellites provide us with a comprehensive set of images showing us what Earth looked like going back to the early 1970’s, elevation datasets often have decade-long gaps between them. As an example, the Shuttle Radar Topography Mission (SRTM) is a global elevation dataset that was collected in 2000; the next global elevation dataset is the ASTER Global Digital Elevation Model which was collected in 2009. In addition to understanding what mined landscapes look like today, it is also important to have a baseline dataset that shows what the region looked like before mining operations began. We recently acquired a historic elevation dataset, courtesy of the USGS, that provides information about what the Central Appalachian landscape looked like in the 1960s and 1970s. By comparing historic and contemporary data, we will gain a better understanding of the nuances of mining operations: by identifying areas that experience loss of elevation we can better detect the areas cleared for mining; by identifying areas which experience elevation gain, we can highlight valley fills. Having a clearer understanding of where each aspect of MTM takes place will facilitate better research into the different environmental impacts resulting from each practice.
The images above show the elevation profile of the Hobet mine in 2000 (top) and 2015 (bottom). By compiling elevation data collected at different times we can see evidence of where ridgelines are destroyed and where valleys are filled with overburden.
What’s next?
In the year to come, we plan to continue developing our datasets and analyses to provide researchers and environmental advocates with actionable information about the state of mined lands in Appalachia. Building on our expertise we plan to automate our landscape recovery assessment ensuring that the health of mined lands is regularly monitored and that the information is readily available. We plan to make use of both historic satellite imagery and elevation data to improve our mine detection algorithm, increasing our overall accuracy, and allowing us to determine what pre-mining landscapes looked like. This assessment of historic landscapes will provide essential information that can be used to compare pre- and post-mining ecosystem health.
Each year that passes we gain a better understanding of how mining has impacted Appalachian ecosystems and communities. As we work to improve the uses of our data, we want to ensure that it remains helpful to researchers, environmental advocates, and communities impacted by mining. If you have insights about improvements we can make, ideas for new work that would be useful, or are interested in helping us illuminate these impacts we invite you to reach out and get involved!
Mined the Overstep
/in Internship, Mountaintop Mining /by Rachel PierceIntern Rachel Pierce reports on her work at SkyTruth.
Prospecting
I first heard of SkyTruth at least a decade ago. I was amazed! Like the Eye of Sauron (but for doing good), they monitored polluters and assessed biodiversity. All right around the corner from my classes in that tiny place called Shepherdstown. I just knew: one day I had to get in. Well, my fellow earthlings, this eagle has landed. With a Remote Sensing course under my belt, among other graduate-level GIS coursework, I felt more driven than ever to be a part of this eclectic crew. The unique personalities and semi-casual, remote work atmosphere contrast sharply with the grim reality of why we do what we do.
We can’t do anything to return Appalachia to its original splendor, but we are doing everything we can to improve how we monitor the extent of surface mining as well as post-mining performance. I was excited to collaborate with SkyTruth’s Geospatial Engineer, Christian, and fellow intern Ethan, on how to approach automated detection methodologies.
As you may already know from our previous work, surface mining — and Mountain Top Mining (MTM) specifically — is the most prevalent method of coal extraction in Central Appalachia. It comes with its own history that’s as twisted as the local bedrock. Over the past two years we worked to examine the complex legacy of environmental implications of MTM, in collaboration with Appalachian Voices and which we explored further with our latest publication in Restoration Ecology. We also continue to publish annual updates to our MTM dataset to continue highlighting the ongoing impacts mining has on Appalachia.
As a part of our commitment to highlighting the impacts of MTM, we are developing new approaches to improve the overall performance and accuracy of our detection model. We also want to find out to what extent mines may operate outside of their permitted boundaries. Considering these issues, we are exploring the potential of how utilizing available permit data can improve our overall detections, while also highlighting areas of concern.
Digging Into the Issues
To address the issue of identifying areas of concern, we can frame the problem around the objective: To find out where coal mines may be overstepping designated permit boundaries and to evaluate the area of overstep compared to the accepted tolerance. This complex situation involves a combination of legal, environmental, and geographical assessments:
Reviewing permit documentation and utilizing spatial data in concert with satellite imagery is a practical and useful method for evaluating the extent of any overstep compared to accepted tolerance, as a similar methodology was implemented in the aforementioned post-mining reforestation assessment. Because we can filter active, inactive, or another mine status, we will continue to use coal surface mine permit boundaries to bolster what we hope will become a ML pixel classification training dataset.
False Positives are an issue because they depict areas that exhibit features easily mistaken for an active surface mine. Types of false positive detections include unreclaimed sites, such as where mining activity has recently ceased, artifacts that appear in imagery despite efforts to filter them out completely, and barren landscapes, like open rock face, eroded soils, or even seasonal changes in greenness.
Shortfalls in spatial data collection are every data hunter’s nightmare and are ever-present in our world, despite modern marvels such as the Internet of Things and Software as a Service geospatially linking us all. Completeness, accuracy, and even accessibility are still very real issues that limit the ability of analysis to have trustworthy and impactful results. This is one of the biggest hurdles in conservation work, often paired with time, funding, and ground-truthing constraints.
Alas, we persevere.
Finding out the accepted overstep tolerance for each state is not so easy either. Without any standard, quantifiable limits, any surface mine detections that occur outside permitted bounds will need to be manually scrutinized. The area of overstep can be totaled and we can look for potential trends or patterns in the data. Opening lines of communication with relevant organizations and agency representatives can help bridge this knowledge gap, provided the information is available and accurate.
Maintaining data integrity is important at every step, so it’s paramount that we preserve the original data. While permit data can be sourced from a single entity — such as through the Geomine application from the United States Office of Surface Mine and Reclamation — it is better to go directly to the source, meaning we run into a lack of data standards and continuity across the board. The reasoning is that each Source Agency will have the most up-to-date records, likely with regularly scheduled updates. For example, Virginia’s Department of Energy makes daily updates to their online datastores, and mine permit data is updated regularly throughout a mine’s life cycle (Daniel Kestner, Personal Communication, Dec 2023). The Geomine data hub happens to contain Virginia’s coal surface mine permit boundaries in their composite dataset; however, during our comparison of current permit bounds to prior detections, we found a considerable gap in this data.
To ensure our newly acquired data maintains its integrity, we must create a new ‘normalized’ dataset of all the permit boundaries within the study area. In doing so, we will be able to refer to the data should something come up during extracting and transforming the permit boundary datasets. This can easily be done by selecting relevant records and exporting them to a new file within a database or directory. This process with the normalization, is known as Extract, Transform, and Load (ETL). We select and export (extract), add new attribute fields and populate them (transform) and, finally, export the data into a repository (load).
But what does ‘normalized’ mean, you ask?
Chipping Away at the Goods
Each state has its own set of procedures and naming conventions when it comes to generating spatial data such as permit boundaries. As a mine’s life cycle is dynamic, there are likewise many attributes describing the mine, its status, its type, and so on. But because there are no standardized methods of attributing a mine or its permit boundary, we need to create a dataset with consistent attributes. This is called Normalizing the Data. This means we will look at all the separate permit data (each state’s permit boundary polygons) and create new attributes that describe the permits: State Name, Permit ID, Permit Status, Permittee, Mine Name, Mine Operator, and Source Agency. To populate these new attributes, we can fill in things like State Name and Source Agency, then we’ll use Structured Query Language (SQL) to write expressions that reference the raw data and fill in the corresponding blanks of our new attributes.
Each state’s department of mining delegates different naming conventions for permit statuses, and some states, such as WV, go into excruciating detail while others maintain simpler designations. Alternatively, Virginia maintains separate and distinct datasets on permits with released status, bond forfeitures, and bond statuses. All considered, simplifying permit status to ‘Active,” ‘Inactive,” and ‘Other’ will suffice. Along the way there are bound to be bottlenecks and roadblocks. For instance, the question arose regarding whether a mine’s status has the same technical and legal meaning as permit status. How does one approach this without losing valuable resources? Sometimes an approach is best taken while noting these considerations for future work and continuing as planned. Other times, it means starting over at square one. Careful assessment of each attribute and its definition is necessary early in the process here.
Once we have our new attribute columns in place, and relevant data entered or copied into the empty values, we can finally merge the cleaned, normalized data into a singular unit, where a user can filter permit boundaries by any of the new attributes and our new, beautiful, and clean dataset will remain shiny and intact.
The Haul
The hope is this cleaned permit dataset will feed into the work we aim to do in terms of assessing post-mining performance. If we’re to improve the accuracy of future detections, we must assess the false positives in those prior. Inactive mines will be classified as non-mine pixels. The detections will be passed through a vegetation index (the well-known NDVI algorithm) to ensure a lack of greenness as we expect zero vegetation in an active surface mine. By manually reviewing the prior detections for any of the false-positive scenarios discussed in the section above, we can further train a ML model to evaluate detections based on Permit Status (Active, Inactive, Other).
Stewardship
The effort put into cleaning this permit data, and that of a future ML model, will benefit ongoing work in assessing post-mining performance. By improving our detections while monitoring these mining activities, we continue to keep the spotlight on operations that overstep their bounds, raise awareness of the impacts and dangers of MTM, advocate for those most impacted by MTM, and educate the public on how MTM affects those of us downstream.
The work doesn’t stop here. Mining activity is staged in bonds and while the staging isn’t as important for this aspect of the work in improving our detections of active mining sites, bond release status will be important for reclamation assessment. As we move toward new approaches that leverage this new mine permit boundary dataset, it will be made publicly available.
Working with and learning from the incredibly talented, intelligent, diverse group of driven people at SkyTruth is an extraordinary experience, one I’ll carry with me onward and upward. I am proud to yell it from any mountain top, removed or not.
SkyTruth Helps Unveil Scale of Offshore Energy Development
/in Cerulean, General, Oceans, Offshore Drilling /by Mitchelle De LeonA new study published in the journal Nature offers an unprecedented view of previously unmapped industrial use of the ocean and how it is changing.
Global Fishing Watch offshore infrastructure global map
We’re excited to announce the release of a new study published today in the journal Nature offering an unprecedented view of previously unmapped industrial use of the ocean and how it is changing. The groundbreaking research — led by Global Fishing Watch and co-authored by SkyTruth’s geospatial engineer, Christian Thomas — used machine learning and satellite imagery to create the first global map of large vessel traffic and offshore infrastructure, finding a remarkable amount of activity that was previously “dark” to public monitoring systems.
Key findings:
“It’s super exciting to see this come out after so many years of research,” said Thomas. “It’s a huge benefit to have data like this to protect the ocean and address the climate crisis.”
Researchers from Global Fishing Watch, the University of Wisconsin-Madison, Duke University, UC Santa Barbara, and SkyTruth analyzed 2 petabytes of satellite imagery spanning 2017-2021 to detect vessels and offshore infrastructure in coastal waters across six continents where more than three-quarters of industrial activity is concentrated. The team developed three deep-learning neural networks to identify objects within the dataset (which achieved over 97% accuracy) and classify them as infrastructure, fishing vessels, or non-fishing vessels.
The ocean is an important resource: over 1 billion people rely on it as a primary food source and 80% of the world’s traded goods are shipped across it. However, a large portion of sea-faring vessels remain untracked. This is due to vessels not being required to broadcast their coordinates — some also may intentionally obscure or manipulate them, in the case of illegal activity — or their positions being withheld by local governments. Similarly, details regarding the development of offshore infrastructure are typically kept private by many nations, which has led to a lack of accurate information on human presence in the ocean.
At any given moment, on average, 63,000 vessels were detected during the analysis: 42–49% of these were fishing vessels. Of these fishing vessels, around 72–76% were not publicly trackable. At the same time, the researchers identified 28,000 pieces of offshore infrastructure at the end of 2021, with 48% and 38% corresponding to wind energy and oil production, respectively.
The lack of a comprehensive understanding of where activity occurs in the ocean may affect the growth of industries such as offshore wind generation, mining, shipping, and fishing. At the same time, a more detailed map of human activity in the ocean can help to better inform estimates of greenhouse gas emissions and global fishing trends.
“Identifying offshore infrastructure is critical for understanding offshore energy development impacts and trends, and is crucial data for our work to detect marine pollution events and hold responsible parties to account,” said Thomas. “For example, if we want to understand how much offshore oil and gas infrastructure is leaking methane into the atmosphere, it helps first to know where all of these facilities are located [to better target methane-detecting satellites].”
These findings have already been incorporated into Cerulean, SkyTruth’s new global monitoring system that uses satellite imagery to detect oil slicks and their potential sources, as many spills come from oil and gas infrastructure.
Offshore energy development surged during the study period. Oil structures increased by 16 percent, while wind turbines more than doubled. By 2021, turbines outnumbered oil platforms. China’s offshore wind energy had the most striking growth, increasing ninefold from 2017 to 2021.
“A new industrial revolution has been emerging in our seas undetected—until now,” said David Kroodsma, director of research and innovation at Global Fishing Watch and co-lead author of the study. “On land, we have detailed maps of almost every road and building on the planet. In contrast, growth in our ocean has been largely hidden from public view. This study helps eliminate the blind spots and shed light on the breadth and intensity of human activity at sea.”
While not all boats are legally required to broadcast their position, vessels absent from public monitoring systems, often termed “dark vessels,” pose major challenges for protecting and managing natural resources. Researchers found numerous dark fishing vessels inside many marine protected areas, and a high concentration of vessels in many countries’ waters that previously showed little-to-no vessel activity by public monitoring systems.
The study also shows how human activity in the ocean is changing. Coinciding with the COVID-19 pandemic, fishing activity dropped globally by about 12 percent, with an 8 percent decline in China and a 14 percent drop elsewhere. In contrast, transport and energy vessel activity remained stable.
The study highlights the potential of this new technology to tackle climate change. Mapping all vessel traffic will improve estimates of greenhouse gas emissions at sea, while maps of infrastructure can inform wind development or aid in tracking marine degradation caused by oil exploration.
The open data and technology used in the study can help governments, researchers and civil society to identify hotspots of potentially illegal activity, determine where industrial fishing vessels may be encroaching on artisanal fishing grounds, or simply better understand vessel traffic in their waters.
“Overall the study contributes to a better understanding of the cumulative footprint of human activity at sea, giving us more of an ability to monitor and protect this important environmental resource,” said Thomas.
# # #