We're sorry but this page doesn't work properly without JavaScript enabled. Please enable it to continue.
Feedback

2D/3D soil consumption tracking in a marble quarry district

00:00

Formal Metadata

Title
2D/3D soil consumption tracking in a marble quarry district
Title of Series
Number of Parts
351
Author
License
CC Attribution 3.0 Unported:
You are free to use, adapt and copy, distribute and transmit the work or content in adapted or unchanged form for any legal purpose as long as the work is attributed to the author in the manner specified by the author or licensor.
Identifiers
Publisher
Release Date
Language
Production Year2022

Content Metadata

Subject Area
Genre
Abstract
Complex quarry districts like Apuan Alps’ marble quarries require remotely sensed high resolution data for soil consumption monitoring over the years: extractive activities lead to environmental challenges that require accurate environmental controls issued by the Tuscan Regional Environmental Agency (ARPAT). The Regional Environmental Information System office (SIRA) over the last 5 and 10 years has developed methods and techniques suitable for both 2D and 3D soil consumption monitoring by using free aerial and satellite images and Open Source Geo-spatial Software for data processing and data dissemination useful in controls’ planning and management. Aerial images and LiDAR acquisition, satellite data, RPAS acquisitions have been tested in order to evaluate their suitability in deriving both 2D and 3D indicators with proper resolution to address required spatial-temporal constraints, i.e. yearly monitoring of high resolution changes (spatial resolution between 50cm and 1m). Due to the size of the Area of Interest (AOI) of the Carrara basins, up to 2.5km x 2.5km, stereo satellite and aerial images can be used to obtain precise terrain models by photogrammetric reconstruction useful in 3D soil consumption monitoring, while middle-resolution (10m) multi-spectral satellite images and high-resolution aerial images (50cm-1m) can be used in 2D soil consumption monitoring and quarries’ area regulations by public bodies (natural soil loss, exhausted areas restorations, debris removals and new disposals). Open-access Sentinel-2 multi-spectral satellite images with 10m of spatial resolution have been used to assess coverage changes; the results have been subsequently refined by manual interpretation over 5 years (2016-2021). Both semi-automatic methods based on spectral distances and machine learning techniques have been used to identify areas affected by extraction activities in QGIS 3.x environment over Sentinel-2 images. Free OGC Web Map Services (WMS) made available by the Tuscan Regional Information System have been used to assess changes highlighted by semi-automatic methods: aerial high-resolution images between 2010 and 2019 have been evaluated by visual photointerpretation, allowing to extend to 10 years the 2D soil consumption assessment over the whole area. Comparison of highlighted 2D changes to regulated areas like mapped debris disposals and quarries’ property limits have been used to check proper developments of extraction activities and proper environmental debris management. In turn, 3D changes have been tracked by comparison of 2009 and 2017 free aerial LiDAR data made available for download by the Tuscan Regional Information System, integrated with two stereo models obtained from 2020 and 2022 Pléiades satellite high resolution images (new acquisitions) freely granted by ESA following Project Proposal id 61779 (“Quarry activity monitoring in Apuan Alps”). Stereo satellite B/W images with 50cm of spatial resolution have been processed by using Open Source stereo processing pipelines in Docker virtual environments, obtaining high precision digital surface models (height precision around 1m) after vegetation filtering. 3D changes detected over the years by elevation algebraic comparison, performed in QGIS 3.x environment, highlight quarries characterized by intense extraction activities (extracted marble blocks, characterized by positive quotas differences) and quarry area management (debris disposing and service infrastructure building, characterized by negative quotas differences). The combined usage of both 2D and 3D changes’ indicators can be challenging in term of proper representation of soil consumption dynamics over the years: while decision makers need a quick and easy access to both 2D and 3D data, web technologies suitable for a proper representation have been developed in very different contexts, making their integration quite complex. While a ‘classical’ 2D webgis client Openlayers or Leaflet-based can be enough to highlight 2D changes and – with some limitations – 3D changes as elevation differences, a ‘true’ 3D visualization environment must be set to track ongoing extraction activities aiming to assess both (a) compliance to authorized extraction plans by public bodies and (b) proper debris management in quarry areas. In addition, 3D web viewers are mainly targeted to represents point clouds or CAD drawings, making very difficult the integration of 2D, 2.5D (Terrain Models) and 3D (extracted volumes) data. A dual 2D/3D webgis client have been developed for proper representation of 2D/3D spatial indicators of ongoing extraction activities in the Carrara marble basin: high resolution images have been served as tiled data, while 2D/3D spatial indicators are served as static and/or tiled vector data. Open-Source libraries have used in data processing, serving and representation inside a map interface. For each quarry included in the Carrara basing, both area limitations and authorized areas for extraction activities have been superimposed over the spatial indicator layers, thus allowing users to easily locate areas subjected to intense extraction activities and to evaluate compliance to sustainability plans and environmental management prescriptions issued by public bodies. 2D and 3D indicators are in progress to be used in prioritizing environmental controls’ planning: this novel application would require a proper scoring system based on the degree of compliance to both environmental management prescriptions and performances mainly in the field of quarry and marble slurry waste management.
Keywords
Block (periodic table)Surface of revolutionExplosionMassDivisorSurfaceDilution (equation)Stochastic processRun-time systemVector potentialPerformance appraisalGUI widgetData managementVolumeNatural numberCore dumpRootReduction of orderGame controllerInsertion lossQuery languageCore dumpProcedural programmingMathematicsVideo gameRun-time systemNatural numberBitProduct (business)Volume (thermodynamics)Forcing (mathematics)Point (geometry)DivisorComplex (psychology)Order (biology)ResultantAreaVector potentialDialectPerformance appraisalForm (programming)Covering spaceWebsiteKnowledge representation and reasoningComputer animation
Limit (category theory)Reduction of orderVolumeSurfaceMathematicsInsertion lossTerm (mathematics)AreaCore dumpSurfaceAssociative propertyComputer animation
Limit (category theory)SatelliteLimit (category theory)Natural numberCore dumpAreaMathematicsReduction of orderCASE <Informatik>WebsiteQuery languageOpen setComputer animation
WebsiteQuery languagePrincipal idealCASE <Informatik>Core dumpWebsiteMathematicsControl engineeringFrequencyComputer animation
Texture mappingFormal verificationOpen sourceOpen setProcess (computing)SatelliteMathematical analysisSeries (mathematics)Computer animation
Texture mappingFormal verificationSatelliteMathematical analysisProcess (computing)Open sourceOpen setCovering spaceInsertion lossNatural numberTotal S.A.Core dumpArchaeological field surveyApplication service providerVolumeSample (statistics)Raw image formatBlock (periodic table)Product (business)Route of administrationView (database)Assembly languageTransport Layer SecuritySurfaceMathematical optimizationMathematical modelStochastic processService (economics)Instance (computer science)MathematicsCuboidNatural numberDigital photographyCore dumpMetreQuery languageAreaTwitterLevel (video gaming)WebsiteVisualization (computer graphics)Web 2.0CASE <Informatik>Mathematical modelMedical imagingProduct (business)Multiplication signPhysical systemElectronic data processingData managementDeclarative programmingImage resolutionControl engineeringOrder (biology)Texture mappingPoint cloudAngular resolutionMathematical modelMathematical optimizationProcess (computing)SatelliteMathematical analysisSoftwareSurfaceClosed setInterpreter (computing)Resultant1 (number)Traffic reportingData structureComputer animation
Polygon meshBuildingData compressionSmoothingServer (computing)TesselationWeb 2.0Point cloudRun-time systemPolygon meshPoint (geometry)Computer animation
MathematicsCovering spaceInsertion lossNatural numberVolumeCore dumpMathematical modelSet (mathematics)Representation (politics)Open setRun-time systemSatelliteAngular resolutionOpen sourceProcess (computing)Data managementCapability Maturity ModelPressure volume diagramScale (map)Angular resolutionCovering spaceKnowledge representation and reasoningRun-time systemMathematicsMathematical modelMathematical modelSatelliteGraph coloringQuery languageWeb pageAreaComplex (psychology)Open sourceRepresentation (politics)Volume (thermodynamics)Process (computing)Order (biology)Office suiteMeasurementAssociative propertyNatural numberWordGroup actionComputer animation
Transcript: English(auto-generated)
Thank you very much. I'll point you to one. In this talk, I will explain the results of the 2T and 3D environmental monitoring in a complex marble query district, the Carrara one, which is a very famous extractive basin.
And we start with just a little of history in order to have a starting knowledge
of environmental challenges we are facing. After this, we will go deeper inside the 2D and 3D monitoring.
And finally, we will see some force solution for changes representations.
And a bit of history of the Japon Alps district. The Japon Alps are a big extractive basin. Since our own age, marble was extracted in this area. And the big changes rose in the first year of the 18th century
with a dramatic increase in marble production. And this dramatic increase led to many problems.
The main factor of environmental pollution is the presence of very big waste dumps coming from past structural works.
And with rainfall, these great dumps become marble slurry that go into the riverbeds,
so disrupting every form of life. And so this is a very important environmental question. And the Tuscany region has issued a special monitoring
project that involved many directorates and involved with us as also the regional environmental protection agency. And the project goals are mainly the enforcing
of in-situ environmental controls. And especially, and this is regarding us, the evaluation of the potential of the remote sensing techniques in environmental controls.
Prioritations by assessing relevant land and COVID changes and instructed in-situ disposal volume changes. I mean, rocks extracted from a query. And in the 2D land COVID change assessment,
we looked for this, for kind of changes in natural soil loss, that is a loss of vegetation
by opening of new queries or the disposal of new dumps, on-site dumps removal, since some dumps can be removed
for new extraction activities. And it's something we're really feeling, since sometime it's a very expensive procedure to dispose extraction dumps outside the extraction site.
This is still a quite used technique. And another kind of oil and COVID changes is natural vegetation growth over exhausted areas.
But why we are assessing for precisely these four kinds of land COVID changes. These changes are important since they can be related to sustainability goals.
In particular, the 2D changes can be associated to monitoring of protected areas. And surface changes also are important in terms
of assessment of carbon stock coming from the loss of vegetation. The extraction dump problematic is relevant.
And it can be assessed with not only 2D, but also with 3D changes detection. Here we have some example of this kind of changes.
We have the border of natural area, which is interested by extraction activity and dump disposal. So here we have a case of no respect
for the protected area limits. In the first case, we have a piece of natural rock that in 10 years have been covered by waste dumps.
On-site dump reduction also can occur by opening of new queries. And this is a case of a query located far outside
the principal roads. So that removing of extraction disposal is not economically justified.
So that we can see in changes between 2012 and 2017 that we have many dumps on site.
And also the same thing can be seen on the other control period, 2017 and 2020.
As for the 2D monitoring, we are fortunate. We are lucky since we have a big series with very close
time intervals of various photos. We have photos at high resolution, less than one meter, for instance, every three years. And we can integrate this service with satellite ones.
In particular, in this work, we focus on the 50 centimeter stereo images in order to derive a DSM, a three stereo DSM.
For software usage, it was used for manual digitizing and for auto-rectifying the player's images,
texture mapping with an analysis with a toolbox, self-processing with a snap or a toolbox snap allows us to have auxiliary layers to improve
the quality of visual interpretation. The results are here. We have a layer with the surface changes over here
and a layer with the kind of changes we detected. And we can see that the trend on natural soil loss is constant over the last two years,
while the other trends are increasing for damper removal and when the query refilling. As for 3D monitoring, we have a wider time interval
since we only have two light aerial service in 2012 and 2017 integrated with the two stereo 3D models.
In this case, we use two pipelines. We tested them. The first is the NASA Ames Pipeline and the other, the S2P, and the self-processing also enabled us
to have an auxiliary layer, but it's less precise since we use the cosmos climate image with three meters of resolution.
And the 3D changes compared with reported production by query owners allows us to find some particular case
of, we can say, some suspected declaration by query manager. In this case, for instance, we have only production data and no in situ waste.
This is very, very strange, but it can be due to reusing of this waste on site for the structure for service infrastructure,
but this case require an additional control by our technicians. Well, let's go ahead to the publishing of 3D data and we have some detailed models at the query level
and some models at the basic level. So we are building a 2D map to access these two kinds of data
and data processing of 3D changes can be very challenging. We use the PIDAL tool.
The PIDAL tool allows starting from point cloud to obtain mesh. As for the first aerial LIDAR, we have to color it
with the closer available aerial imagery. After this, we have to clip them over each basin area and transform the proper coordinate system,
assembling the two models allows us, by passing between the optimization stage, to obtain a model that can be published on the web. You know that when publishing 3D models on the web,
we have to keep its size small. And starting from point clouds, we have a single tile
that is about 32 megabytes. After reduction, after meshing, simplifying and compressing, we obtain a single basin that can be published in 3D hope environment
that is about 25 megabytes. And this is a compressed mesh that can be served by a smooth web server. So we have produced this kind of model. The first model, that is a natural color model.
The second one overlaid with the 2D layer of land cover changes. And another textured with extracted volumes and new dams.
And so we are assembling these models on a single page that allows the user to select the model and visualize it.
This can be an affordable tool for our technicians in order to check the query associated with measured impacts. Concluding both 2D and 3D aerial and satellite resolution data,
allow environmental monitoring of complex extractive basins and representation in 3D of larger amounts of data
can be challenging, but open source tools and processing are ready for achieving the goal of enabling complex representations.
And this combined representation for us is a must have to grant a quick look to our technicians of the main changes occurred in the query area.
And this work has been made by a working group, including other four colleagues of my office.
Thank you very much, Cynthia.