Skip to content

Risk Academy, led by the Nexus Consortium · Risk Science and Methods · Department of Geospatial and Risk Data Science

GIS, remote sensing, exposure mapping, GeoAI and the accuracy of what a map shows

भू-अवकाशीय विश्लेषण, पृथ्वी निरीक्षण आणि GeoAI

A map is a model with an accuracy that can be measured.

Map geospatial analysis, Earth observation and GeoAI on a real placeOpen RAT-201

Illustrative: layers of units and their weighted connections, an activation front sweeping from input to output

01व्याप्ती

What the discipline covers

The subtopics its courses and labs teach.

अभ्यासक्रम
1 course open
स्तर
L1 Introductory, L2 Intermediate, L3 Advanced
प्रयोगशाळा
Exposure lab, GeoAI lab, Map lab, Geo quiz
  • GIS foundations
  • Remote sensing and image analysis
  • Radar flood mapping
  • Exposure mapping
  • GeoAI methods
  • Map accuracy and validation
  • Spatial statistics
  • Cartography for risk
  • Earth observation sensors and missions
  • Disaster monitoring from space

02नकाशे

Build a twin of any place

Search any place or open the world; the layers and generators follow geospatial analysis, Earth observation and GeoAI on the planet map, the globe and in 3D.

The central Apennines: L'Aquila, Amatrice and Norcia: a flat plan of the faculty twin. Terrain 24 to 2,805 m; 3 reference earthquakes, M6.2 to M6.6; Earthquakes M4.5 and above, past week: read live when the 3D twin opens; USGS ShakeMap intensity contours: read live when the 3D twin opens. Contours every 500 metres from the elevation grid; reference earthquakes as marks.
The central Apennines: L'Aquila, Amatrice and Norcia, Italy. Flat plan of the same data: relief, the three reference epicentres with magnitude and depth, and the numbers below; the sampler's figure shows one precomputed field with its parameters. The 3D twin opens on devices that can draw it, once its data and code have been checked.
  • Terrain 24 to 2,805 m
  • 3 reference earthquakes, M6.2 to M6.6
  • Earthquakes M4.5 and above, past week: read live when the 3D twin opens
  • USGS ShakeMap intensity contours: read live when the 3D twin opens
Credits and dates per layer are listed under the stage. Relief in the twin is exaggerated ×1.5 for legibility.

Drag to move, right drag or Shift with a drag to turn and tilt, Ctrl with the wheel to zoom. Keyboard: arrows move, Shift with arrows turn and tilt, + and – zoom, 0 resets, V steps through the views, Escape clears.

Flat plan of the same data: relief, the three reference epicentres with magnitude and depth, and the numbers below; the sampler's figure shows one precomputed field with its parameters. The 3D twin opens on devices that can draw it, once its data and code have been checked.

The boundaries, names and designations shown do not imply any judgement on the legal status of any territory, or any endorsement or acceptance of such boundaries. Countries: Natural Earth (public domain).

03The subject

The subject, methods and measures and debates and bodies

The discipline

The field and its scope

For analysts, modellers and mapmakers who use geographic information, satellite observation and spatial machine learning to locate hazards, exposure and damage, and who must judge what a map can and cannot show.

Geospatial analysis, Earth observation and GeoAI study how risk is located: where hazards occur, what is exposed, how damage spreads and how all of that is measured from space, from the ground and from models. The field combines geographic information science (coordinate systems, data models, spatial analysis), remote sensing (optical and radar sensors, image classification, change detection), spatial statistics (autocorrelation, interpolation, scale effects) and machine learning on spatial data, now including foundation models pre-trained on petabytes of satellite imagery. Its users are emergency mapping services, catastrophe modellers, planners, insurers and humanitarian agencies.

Practice is organised around the data chain from acquisition to decision. Open missions (Landsat, Sentinel-1 and Sentinel-2, MODIS) and commercial constellations supply imagery; cloud platforms process it at continental scale; products such as flood extent, building footprints and settlement layers feed exposure and damage estimates; cartography carries the result to users. Accuracy assessment with designed samples and the critical reading of what a map claims are the discipline's quality control. Its open questions concern the transfer of learned models across regions, the validation of products where ground truth is scarce, and the ethics of mapping people from above.

Measures

Quantities and units

QuantityUnit or scaleWhat to know
Positional accuracyMetres, as root mean square error against control pointsHow well a feature's mapped location matches the ground. Published satellite products quote accuracies of a few metres to tens of metres; the figure applies to the control sample, not to every pixel.
Thematic accuracy (overall accuracy, producer's and user's accuracy, F1)Proportions, dimensionless, from a confusion matrix on a probability sampleHow often the mapped class is right. Olofsson and co-authors (2014) set out good practice: a designed sample, stratified estimates with confidence intervals and area estimates adjusted for the confusion matrix. Accuracies reported on training data or convenience samples are inflated.
Spatial resolution and revisit timeMetres per pixel; days between acquisitionsThe limits of what a sensor can see and how often. Sentinel-2 resolves 10 metres every five days; buildings and vehicles need sub-metre commercial imagery with less frequent coverage.
Flood extent and depthSquare kilometres; metresRadar mapping of open water is reliable; urban, vegetated and shallow flooding is under-detected. Depth requires a terrain model and carries its vertical error, typically a metre or more for global models.
अवकाशीय स्वसहसंबंधMoran's I or local indicators, dimensionlessThe degree to which nearby values are similar. It invalidates independence assumptions in statistics and random train-test splits in machine learning, and it is the reason spatial blocking is needed.
Model transfer performanceAccuracy or F1 on a held-out region or period compared with the training regionHow well a learned map generalises. Drops of ten to thirty percentage points across regions are common and are the honest figure for operational use elsewhere.

पद्धती

Methods and their limits

Each method is listed with what it assumes and where it fails.

पद्धतAssumesFails when
Supervised image classification and segmentationThat labelled training areas represent the classes across the image, that spectral or textural signatures separate them, and that labels are accurate.Across seasons, sensors and regions without retraining, for classes that look alike (bare soil and demolished buildings), and when labels come from the same map the model is meant to improve.
Radar flood mapping (Sentinel-1 change detection and thresholding)That open water appears dark in synthetic aperture radar, that a pre-event image is available, and that terrain shadow and wind-roughened water can be masked.In cities (double bounce), under vegetation, for shallow or short-lived floods between overpasses, and when the terrain mask is coarse.
Exposure mapping from footprints, settlement layers and census disaggregationThat building footprints and population can be derived or disaggregated to grid cells with known error, and that attributes can be inferred from form and context.Where footprints are missing or merged, where dasymetric assumptions put people in the wrong places, or where the attributes that matter for loss (construction, use) are not visible from above.
Spatial statistics and interpolation (kriging, local indicators, regression with spatial terms)That the variable is spatially continuous with a stationary covariance structure, that the sample design is known, and that the scale of analysis matches the process.Under non-stationarity, across the modifiable areal unit problem when results change with the zoning, and when interpolation crosses barriers (rivers, ridges) that the model ignores.
Geospatial foundation models and satellite embeddingsThat representations learned from large unlabelled archives transfer to downstream tasks with few labels and that embeddings preserve the features relevant to the task.When the task depends on information absent at the pre-training resolution, when embeddings encode regional artefacts, or when a decline in performance in low-data regions goes unmeasured because there are no labels to measure it.
Accuracy assessment with designed probability samplesThat an independent reference sample with a known inclusion probability can be collected and interpreted consistently by analysts.When reference data are themselves derived from imagery of the same date and resolution, when the sample is too small for rare classes, or when the assessment is skipped because fieldwork is costly.

Debates

What the field debates

Do geospatial foundation models change what can be mapped, or only how cheaply?
Proponents cite benchmark gains with few labels, continental embeddings that allow similarity search and change detection, and the release of open models such as Prithvi. Sceptics observe that gains are smallest for tasks that matter most for risk (damage grades, informal settlements), that evaluations use benchmark regions, and that the models' training data are not documented well enough to say where they will fail.
Global products or local mapping?
Global flood, settlement and footprint layers are defended for consistency, coverage of places with no local capacity and reproducibility, and their stewards publish accuracy statements. Local mappers reply that global products carry large local errors (Tellman and co-authors report regional under-detection), that they are used without reading the accuracy statement, and that local knowledge of drainage, construction and land use is not recoverable from pixels.
What are the limits of mapping people from above?
One position holds that counting buildings and people from imagery is necessary for exposure and humanitarian response and that resolution limits protect individuals. The other points to the identification of settlements, religious or ethnic patterns and displacement from imagery, argues that consent and benefit to the mapped are absent, and calls for the data governance principles of CARE and for impact assessments before release.

Bodies and data

Bodies that set practice, and the data they publish

Bodies and standards are named so learners can find and apply their published work; the Academy teaches their use and links to the publisher, and no affiliation is implied.

BodyRole in the field
Open Geospatial Consortium (OGC)Publishes the open standards for geospatial data and services (OGC API, GeoTIFF, GeoPackage) and co-develops the SpatioTemporal Asset Catalog community specification.
ISO/TC 211 Geographic information/GeomaticsMaintains the ISO 19100 series on geographic metadata (ISO 19115), data quality (ISO 19157) and reference systems.
Committee on Earth Observation Satellites (CEOS) and the Group on Earth Observations (GEO)Coordinate the space agencies' observation programmes, calibration and validation, and the data sharing principles behind open imagery.
European Space Agency and the European Commission Copernicus programmeOperate the Sentinel missions and the Copernicus Emergency Management Service that provides rapid mapping after disasters.
United States Geological Survey and NASAOperate Landsat and the Earth science missions, maintain the open archives and co-develop the open geospatial foundation models.
IEEE Geoscience and Remote Sensing Society and the International Society for Photogrammetry and Remote SensingThe learned societies for remote sensing method, benchmarks and accuracy assessment practice.
DatasetSteward
Copernicus Data Space Ecosystem (Sentinel archives)European Space Agency for the European Commission
Landsat missions archiveUnited States Geological Survey
Global Human Settlement LayerEuropean Commission Joint Research Centre
OpenStreetMapOpenStreetMap Foundation
Copernicus Emergency Management Service rapid mappingEuropean Commission Joint Research Centre

04पद्धत

From a radar image to the people in the water: a flood map, its accuracy sample and the area the sample corrects

A radar scene is classified into flooded and dry pixels; a probability sample of reference points gives the map's user's and producer's accuracy; those two rates correct the mapped area into an unbiased estimate with a confidence interval, and the corrected area over a population layer gives the exposed count a response decision can use.

Sources: Olofsson et al., Good practices for estimating area and assessing accuracy of land change, Remote Sensing of Environment (2014); Tellman et al., Satellite imaging reveals increased proportion of population exposed to floods, Nature (2021); Torres et al., GMES Sentinel-1 mission, Remote Sensing of Environment (2012); Zhu et al., Deep learning in remote sensing: a comprehensive review and list of resources, IEEE Geoscience and Remote Sensing Magazine (2017); Tobler, A computer movie simulating urban growth in the Detroit region, Economic Geography (1970).

05At a glance

At a glance, with 5 related topics

3D आलेख

जोखीम विज्ञान आणि पद्धती विद्याशाखा

265 nodes1,691 relations9 views

Explore the 3D graph: Faculty of Risk Science and Methods

06प्रयोगशाळा

How the subject is practised in labs

LabWhat you do in itCourses using it
Exposure labOverlay people and assets with hazard footprints5
GeoAI labApply Earth observation and machine learning to open imagery4
Map labRead hazard, exposure and boundaries on real maps3
Geo quizCheck spatial reasoning on where hazards and systems meet3

07नियंत्रण कक्ष

How Geospatial analysis, Earth observation and GeoAI is taught, as a process you can run

Press Play to run this process from this page's own data.

भू-अवकाशीय विश्लेषण, पृथ्वी निरीक्षण आणि GeoAI

How Geospatial analysis, Earth observation and GeoAI is taught

5 subjects in scope, 3 key terms, 1 course open, 4 labs, in the Geospatial and Risk Data Science department of Risk Science and Methods: from the scope to the method and its terms, the courses that teach it, the labs on the twin, the assessment and the record.

  1. 01 Scope

    1. 01

      GIS foundations

      Start · the discipline

      GIS foundations

    2. 02

      Remote sensing and…

      Start · the discipline

      Remote sensing and image analysis

    3. 03

      Radar flood mapping

      Start · the discipline

      Radar flood mapping

    4. 04

      Exposure mapping

      Start · the discipline

      Exposure mapping

    5. 05

      GeoAI methods

      Start · the discipline

      GeoAI methods

  2. 02 Method and terms

    1. 09

      From a radar image to…

      Model · the learner

      A radar scene is classified into flooded and dry pixels; a probability sample of reference points gives the map's user's and producer's…

    2. 10

      Raster and vector data

      Record · the discipline

      Grids of cells and geometric features, the two main forms of spatial data.

    3. 12

      Accuracy assessment

      Record · the discipline

      Comparing a map with reference data to estimate its errors.

    4. 14

      Coordinate reference…

      Record · the discipline

      The definition that ties coordinates to locations on the Earth.

  3. 03 Courses

    1. 11

      RAT-201 EO/GIS Starter…

      Step · the learner

      EO/GIS Starter Lab: Maps, satellites, terrain and open data for hazard and exposure analysis, with four hands-on labs

  4. 04 Labs

    1. 06

      Geoai

      Exercise · the learner, on the faculty twin

      The method runs on the faculty twin and the admitted map layers.

    2. 07

      संपर्कात येणे

      Exercise · the learner, on the faculty twin

      The method runs on the faculty twin and the admitted map layers.

    3. 08

      Geoquiz

      Exercise · the learner, on the faculty twin

      The method runs on the faculty twin and the admitted map layers.

    4. 13

      जुळवा

      Exercise · the learner, on the faculty twin

      The method runs on the faculty twin and the admitted map layers.

  5. 05 Assess

    1. 15

      Module checks

      Gate · the course

      Each module ends with a short check against its outcomes.

    2. 16

      Assessed work

      Decision · the learner

      The assessed work that closes a course: a lab record, a case memo or an assignment, reviewed against the rubric.

  6. 06 Record

    1. 17

      पूर्णता नोंद

      Outcome · the learner

      The record stacks into the discipline achievement.

    2. 18

      Discipline achievement

      Record · the Academy

      The discipline achievement follows the discipline rule, read from the records.

Geospatial analysis, Earth observation and GeoAI · Geospatial and Risk Data Science An outline of how the work is designed to run; it does not report the status of any decision or record.

08शिक्षण

Learning, practice and connections

Achievements

Achievements in this discipline

  1. Course achievements

    One for each course with this as its primary discipline, earned when you pass the course's assessed activities. One can be earned now.

  2. Discipline: Geospatial analysis, Earth observation and GeoAI

    Complete 3 assessed courses in geospatial analysis, earth observation and geoai, at least 2 with it as their primary discipline and at least 1 at Intermediate (L2) or above, totalling at least 45 measured learning hours. In development.

  3. Department: Geospatial and Risk Data Science

    Earn two discipline achievements in the Department of Geospatial and Risk Data Science. In development.

  4. Programme: Data and GeoAI specialist

    Complete every core course of the Data and GeoAI specialist under assessed completion, ending with its capstone. In development.

  5. Cross-faculty: Space and food

    Courses in connected disciplines and a passed integrative task. In development.

An achievement is a record of the learning completed and assessed here. How recognition works.

मार्ग

Pathways that teach it

Where this is practised

Where the subject is practised

  • The Global Centre for Risk and Innovation (GCRI Canada)Space and Earth ObservationSpatial temporal evidence, EO in water and agriculture, Earth observation sensors, Disaster monitoring
  • FoundryMaps and GIS

Field platforms of the ecosystem where the subject is studied and applied.

Practised in

The two Foundry systems where it is practised

Foundry system pages describe what each system and its rooms are designed to do; the courses teach the methods these systems apply.

Connected disciplines

Disciplines it depends on

Across the ecosystem

Across the ecosystem

The science, platforms and standards this discipline draws on and contributes to across the ecosystem.

Foundations in the STEM Commons

Where the field is worked on across the network

  • The Global Centre for Risk and Innovation (GCRI Canada)FoundryThe platform's work on this field.

Standards for the field

Pages of the ecosystem, cited so learners can follow the field into research, platforms, standards and national practice.

09सहभागी व्हा

Take part in Risk Science and Methods

तुम्ही कोण आहात तिथून सुरुवात करा.

Routes

  • मार्ग

    राष्ट्रीय संस्था

    Ministries, statistics offices and agencies: a national desk and chapter, and a seat in the working groups. National chapters

  • मार्ग

    Sectors and professions

    Operators, regulators, banks, insurers and professional bodies whose work touches risk science and methods. For organisations

  • मार्ग

    शिक्षण

    Courses and practice, with a record of every course completed. Free to study. Open learning

  • मार्ग

    मानके

    Draft, review and adopt the methods and data standards with the working groups. Working groups

  • मार्ग

    Communities and research

    Universities, civil society and individual experts: projects, data and review. Write to the Academy

Open calls

Open callSeeksCloses
Founding call for experts to the working groups CALL-FOUNDING-EXPERTSexpert, editor, chair2026-12-15
Open nomination for the three independent expert seats of the Global Standards Council CALL-GSC-INDपरिषद-सदस्य2026-12-01
Nominations of interim regional members (applies only if the transitional provision is decided) CALL-GSC-REGपरिषद-सदस्य2026-12-15
Review of the wave-1 risk core outlines (C1, C2, C3, C4, C8, C10, C11) CALL-REVIEW-WAVE1contributor2027-01-31
जोखीम विज्ञानाच्या मुख्य मानकांसाठी तज्ज्ञ: अनिश्चितता आणि पुरावा श्रेणीकरण, मापनश्रेणी, संरचित तज्ज्ञ निर्णय, सहनक्षमता, टोकाच्या आणि संयुक्त घटना, आपत्तीजनक आणि अस्तित्वाला धोका निर्माण करणारी जोखीम, नीतिमत्ता आणि समता, एकके आणि मापन (NRS-C17 ते NRS-C24) CALL-CORE-SCIENCEexpert, editor, chair2027-01-31
Independent reviewers for the families of the hazard atlas, three from at least two regions for each family (NHS2-R19), priority families first CALL-ATLAS-REVIEWतज्ज्ञ2027-03-31
अंतरिम अग्रगण्य सुरक्षा आढावा मंडळाचे सदस्य आणि रेड-टीम समीक्षक (जैवसुरक्षितता आणि जैवसुरक्षा, AI सुरक्षितता आणि सुरक्षा, महत्त्वाच्या पायाभूत सुविधांची सुरक्षा); अर्ज गोपनीयपणे हाताळले जातात CALL-FRONTIER-SAFETY-REVIEWतज्ज्ञ2026-11-30
आवश्यकता नोंदवही, NHS-2 यंत्र प्रतिनिधित्व, अनुरूपता चाचणी संच आणि अनुरूपता मूल्यांकन योजनेसाठी ऑन्टोलॉजी, मापनशास्त्र, सॉफ्टवेअर पडताळणी आणि अनुरूपता मूल्यांकनातील तज्ज्ञ CALL-CONFORMANCE-ONTOLOGYexpert, editor2027-01-31

Calls past their closing date are hidden automatically. Every call, proposal and route: open areas and calls.

Cells and CERTs

  1. Gather the cell

    Bring together three to seven people with complementary skills: data, field practice, policy and communication.

  2. Register it

    Write to the team with the cell's name, members, area and region. You receive a reference and a contact. Write to the Academy

  3. Choose a charter

    Pick one focus: research and evidence, data and maps, standards review, or community preparedness.

  4. Learn together

    Follow shared study paths at Risk Academy, with practice and a record of every course completed. Hazards and resilience

  5. Lead and report

    Hold regular sessions. Publish what the cell learns through Reports. Bring findings to the working groups. Working groups

  • Specialise as a Community Emergency Response Team (CERT)

    A CERT prepares its neighbourhood for the hazards it faces. It builds a local hazard profile from this site's map. It trains in household and street preparedness. It learns first aid and light search through recognised local providers, and it runs drills. A CERT works alongside the local emergency services and follows their direction. It never replaces them and is not an emergency service.

निर्देशांक

What a national team or a local community can do on The Global Risks Index (NXS-GRX) today.

  • NXS-GRX

    Your country's profile

    Open risk profiles for 195 countries, with sources and licences on every value. Open the Index

  • NXS-GRX

    Correct or reply

    Correct a value or reply for your country. The change is logged with its effect. Open the Index

  • NXS-GRX

    Offer data

    Offer a dataset. It enters through the licence gate, with your credit. Open the Index

  • NXS-GRX

    Use the open data

    Layers, profiles and tiles through the open API. Open the Index

10References

Core references and where to begin

Start with Tobler, then Olofsson, then Anselin.

  1. Tobler, W. R., A computer movie simulating urban growth in the Detroit region, Economic Geography (1970). DOI
  2. Olofsson, P. et al., Good practices for estimating area and assessing accuracy of land change, Remote Sensing of Environment (2014). DOI
  3. Anselin, L., Local indicators of spatial association, LISA, Geographical Analysis (1995). DOI
  4. Torres, R. et al., GMES Sentinel-1 mission, Remote Sensing of Environment (2012). DOI
  5. Pekel, J.-F., Cottam, A., Gorelick, N. and Belward, A. S., High-resolution mapping of global surface water and its long-term changes, Nature (2016). DOI
  6. Gorelick, N. et al., Google Earth Engine: planetary-scale geospatial analysis for everyone, Remote Sensing of Environment (2017). DOI
  7. Zhu, X. X. et al., Deep learning in remote sensing: a comprehensive review and list of resources, IEEE Geoscience and Remote Sensing Magazine (2017). DOI
  8. Janowicz, K., Gao, S., McKenzie, G., Hu, Y. and Bhaduri, B., GeoAI: spatially explicit artificial intelligence techniques for geographic knowledge discovery and beyond, International Journal of Geographical Information Science (2020). DOI
  9. Tellman, B. et al., Satellite imaging reveals increased proportion of population exposed to floods, Nature (2021). DOI
  10. Jakubik, J. et al., Foundation models for generalist geospatial artificial intelligence, arXiv (2023). arxiv.org

11प्रश्न

प्रश्न, उत्तरे आणि प्रमुख संज्ञा

येथे अभ्यास विनामूल्य आहे का?

होय. Nexus Consortium च्या नेतृत्वाखालील Risk Academy चा प्रत्येक अभ्यासक्रम आणि शिक्षण साहित्य विनामूल्य आहे, स्व-गतीने किंवा गटासोबत. सामान्य सहभागासाठी शुल्क नाही; सशुल्क काम, इंटर्नशिप आणि अधिछात्रवृत्ती या स्वतंत्र व्यवस्था आहेत.

अभ्यासक्रम पूर्ण केल्यावर मला काय मिळते?

A completion record: the course, its version, the date and what was assessed, private by default and shared by you. It stacks into discipline, department and programme achievements. A record states what was studied and assessed and does not certify competence; it is not a licence, a professional title or an employment status.

माहिती आणि ठिकाणे कुठून येतात?

Every faculty teaches on a twin of a real place built from open data; each layer carries its provider, licence and date, and every result on the Academy states its source. Maps and figures are indicators for teaching, never an assessment of a place.

How do I take part beyond the courses in geospatial analysis, Earth observation and GeoAI?

Form a competence cell of three to seven people, study a shared pathway, take a supervised assignment in the National Working Groups, or bring a cohort from your institution. For organisations.

प्रमुख संज्ञा

Raster and vector data
Grids of cells and geometric features, the two main forms of spatial data.
Accuracy assessment
Comparing a map with reference data to estimate its errors.
Coordinate reference system
The definition that ties coordinates to locations on the Earth.
GeoAI
The use of machine learning on spatial and Earth observation data.

अहवालयोगदान द्या

या पानावरून पुरावा योगदान द्या.

Reports is the Consortium's quarterly publication. What you know about this subject can become evidence, a case or a correction in the next edition, credited to you and published under CC BY 4.0 where it is your own work.

Risk Academy

Never send personal data, sensitive infrastructure detail or anything you have no right to share. A contribution is reviewed before use; sending it confers no authorship or editorial rights, and sponsorship never buys a finding.

भू-अवकाशीय विश्लेषण, पृथ्वी निरीक्षण आणि GeoAI

Start with an open course

The catalogue lists each course with its outcomes, labs, hours and level.

Open RAT-201Back to Department of Geospatial and Risk Data Science