Each topic below explains the evidence, the business problem it creates, how GLOBEIR solves it, the benefits to your organisation, and how your data stays private and secure.
| Mission |
Sensor |
Key specification |
| Landsat 9 (2021) |
Optical + thermal |
30 m multispectral, 15 m panchromatic, 100 m thermal; 8-day revisit with Landsat 8; part of a record running since the early 1970s [13] |
| Sentinel-2 (ESA Copernicus) |
Optical, 13 bands |
10/20/60 m bands, 290 km swath, 5-day revisit at the equator [14] |
| Sentinel-1 (ESA Copernicus) |
C-band SAR |
Day-or-night, all-weather imaging; 250 km swath in the main mode [15] |
| Sentinel-5P / TROPOMI (2017) |
Atmospheric spectrometer |
Daily global coverage; measures NO₂, O₃, CO, CH₄, SO₂ and aerosols at about 5.5 × 3.5 km [16] |
| Resourcesat-2A (ISRO, 2016) |
Optical |
LISS-4 5.8 m, LISS-3 23.5 m, AWiFS 56 m [17] |
| EOS-04 (ISRO, 2022) |
C-band SAR |
All-weather imaging for agriculture, forestry, soil moisture, hydrology and flood mapping [18] |
| NISAR (NASA–ISRO, launched 30 July 2025) |
L- and S-band SAR |
12-day repeat, 242 km swath [19] |
Spectral indices. NDVI = (NIR − Red)/(NIR + Red). Values around 0.1 or below indicate bare rock, sand or snow, 0.2–0.3 shrub and grassland, and 0.6–0.8 temperate and tropical forest [20]. NDWI = (Green − NIR)/(Green + NIR) highlights open water, though it can overestimate water in built-up areas [21].
Choosing resolution. Coarse sensors (about 1 km to 56 m) cover huge areas often and suit fire alerts, air quality and seasonal crop monitoring. Medium sensors (10–30 m) are the workhorse for land-use change, forest cover and water mapping. Fine sensors (5 m and below) suit site-level work. Optical sensors cannot see through cloud, so SAR is essential for monsoon flood mapping [15][22].
Business problem
Satellite data is free or low-cost [13][14][17], but choosing the right mission, resolution and index is technical, and a wrong choice gives wrong answers: optical imagery is blind under monsoon cloud [15][22], and NDWI can overestimate water in built-up areas [21]. Most ESG teams, project developers and government departments do not have in-house remote-sensing specialists, and reports that do not document their method are hard to defend before a regulator, auditor or lender.
Our solution
- GLOBEIR selects the mission, resolution and index to suit each question, from daily atmospheric data to 5 m site imagery, and combines optical, thermal and SAR data where needed.
- Processing pipelines with cloud masking, documented methods and versioned outputs, so results can be repeated cycle after cycle.
- Field ground truth collected with the My GLOBEIR app to check accuracy.
- Results delivered through interactive WebGIS dashboards and standard GIS files; see our wider services.
Benefits
- Low data cost by building on free Landsat, Sentinel and ISRO archives [13][14][17].
- All-weather continuity, because SAR images day or night through cloud [15][18].
- Long baselines from a Landsat record running since the early 1970s [13].
- No over-spending on very high-resolution data where medium resolution answers the question.
- Documented, repeatable methods that stand up to review.
Privacy & data security
- Data involved: public satellite imagery at these resolutions holds no personal data, but once it is combined with client records such as asset lists, land parcels and geotagged field photos, the combined data becomes sensitive.
- Stored in India: client drone or survey data finer than the DST threshold (1 m horizontal, 3 m vertical) is stored and processed only in India and never sent to non-Indian servers [23][24].
- Encrypted and certified hosting: data is encrypted in transit (TLS 1.2 or higher) and at rest (AES-256), on ISO 27001 / SOC 2-certified cloud infrastructure in India [25].
2. Forest Cover Monitoring
The Forest Survey of India's biennial India State of Forest Report 2023 combines satellite mapping with field inventory: forest and tree cover together total 827,357 sq km, or 25.17% of India's geographical area, up 1,445 sq km since 2021 [26]. FSI's near-real-time forest fire alerts use MODIS (1 km) and VIIRS (375 m) hotspots received at NRSC, processed in Dehradun and sent to registered users by SMS and to forest departments by email, with a filter that removes false alarms from mining and industrial sites [27]. Globally, the Landsat-based study behind Global Forest Watch (now Global Nature Watch [28]) mapped 2.3 million sq km of forest loss and 0.8 million sq km of gain between 2000 and 2012 at 30 m resolution [29].
Forest mapping works because healthy forest canopy gives high NDVI values, typically 0.6–0.8, against 0.2–0.3 for shrub and grassland [20]. Medium-resolution sensors (10–30 m) are the workhorse for forest cover, while coarse sensors suit fire alerts.
Business problem
- Forest departments manage large, remote areas with limited field staff. National fire alerts arrive quickly [27], but turning them into beat-level action, and checking which patches have actually changed, still takes time.
- Plantation companies and afforestation project developers must show that planted areas exist and are surviving. Claims that are not backed by independent, repeatable evidence are hard to defend before auditors, funders or regulators [26][29].
Our solution
- Multi-date classification and change detection from Sentinel-2, Landsat and LISS imagery, with accuracy assessment, to show where forest or plantation cover was gained or lost.
- Fire hotspot data overlaid on range, beat or plantation boundaries, with fire-frequency heatmaps to plan patrols and fire lines.
- Field verification of change and fire points with the My GLOBEIR app.
- Results on WebGIS dashboards for department and project teams.
Benefits
- Evidence built on the same satellite-plus-field approach used for India's national forest accounting [26].
- Change maps at 30 m or finer, the resolution used for global forest-loss mapping (industry example) [29].
- Faster targeting of field effort to the patches and fire points that matter.
- Repeatable monitoring cycles that support plantation progress and afforestation reports.
Privacy & data security
- Data involved: satellite forest maps are non-personal. Field ground-truth photos can capture people or homes in forest-fringe villages, and geotagged photos are personal data under the DPDP Act [30]; a company's plantation boundaries are commercially confidential.
- Stored in India: drone or survey data of plantations finer than the DST threshold (1 m horizontal, 3 m vertical) is stored and processed only in India and never reaches non-Indian servers [23][24].
- Field app controls: background location in the My GLOBEIR app is optional and opt-in with a persistent notification, and location is captured for work purposes only.
- Access control: separate roles for field, supervisor and head-office users, with audit logs of access and changes.
3. Land Use and Land Cover Change
NRSC has produced national land-use maps at 1:250,000 every year since the 2004-05 crop year, and 1:50,000 maps in a 54-class scheme for three time periods [31].
Land-use mapping classifies each pixel into classes such as cropland, built-up, forest and water, then compares dates to show what changed. Sentinel-2 (10 m) and LISS-4 (5.8 m) add site-level detail [14][17], and the Landsat record allows baselines going back to the early 1970s [13].
Business problem
- District administrations and planning departments need to know how cropland, wetlands and open land are turning into built-up area, but national maps at 1:250,000 and 1:50,000 [31] are too coarse for ward, village or project-level decisions.
- Infrastructure and real-estate developers, and ESG teams, need a defensible record of what a site looked like before and after development, often for periods when no survey was done.
Our solution
- District, city or project-level land-use and land-cover classification from Sentinel-2, Landsat and LISS, with change matrices showing what changed, where and when.
- Class schemes that can be aligned with the NRSC classification for comparability with national maps [31].
- Accuracy assessment against field samples collected with the My GLOBEIR app.
- Change layers for district views in administration monitoring and on WebGIS.
Benefits
- Defensible numbers on land-use change, with documented methods and accuracy figures.
- Historical baselines from free archives, without paying for new surveys [13][14][17].
- Local detail at 10 m or finer, where national maps are too coarse [14][17][31].
- One consistent method across districts and over time.
Privacy & data security
- Data involved: land-cover classes are non-personal. Data becomes personal when change maps are overlaid on land parcels linked to owner names [30].
- Minimised data: analysis uses parcel or survey numbers, not owner names, wherever possible, and results can be reported as village or district totals.
- Stored in India: parcel or drone data finer than the DST threshold is created, stored and processed only in India [23][24]; government clients can choose a MeitY-empanelled government cloud [32].
- Encryption: data is encrypted in transit (TLS 1.2 or higher) and at rest (AES-256).
4. Air Quality Monitoring
TROPOMI's NO₂ maps resolve emissions from cities, power plants, industrial complexes, highways and shipping lanes [33], and the measurements feed the Copernicus Atmosphere Monitoring Service's daily air-quality forecasts [33].
Sentinel-5P measures NO₂, O₃, CO, CH₄, SO₂ and aerosols with daily global coverage at about 5.5 × 3.5 km [16][34]. At this coarse resolution it is suited to city and regional trends rather than single chimneys.
Business problem
- City authorities need an area-wide view of pollution trends and seasonal patterns, beyond the readings at individual monitoring points.
- Industrial units, power producers and their ESG teams are increasingly asked about the air-quality context around their sites, but have no independent, consistent picture of how regional pollution has changed over time.
Our solution
- NO₂ and aerosol trend maps from Sentinel-5P around cities, industrial clusters and project sites [16][33].
- Seasonal and year-on-year comparisons, shown as heatmaps and on WebGIS dashboards.
- Clear statements of resolution limits, so results are used for regional trends and not for single-source attribution.
Benefits
- Daily, wall-to-wall observation that ground stations alone cannot give [16].
- Built on the same measurements that feed the Copernicus daily air-quality forecasts (industry example) [33].
- Low data cost, because Sentinel-5P data is open.
- Consistent trend evidence for ESG reports and city planning discussions.
Privacy & data security
- Data involved: atmospheric measurements contain no personal data. The sensitive part is the client's own facility locations, emission records and internal reports, which are commercially confidential.
- Purpose limitation: client data is used only for the agreed purpose; it is not sold or shared, and not used to train models for other clients.
- Stored in India: any site drone or survey data finer than the DST threshold is stored and processed only in India [23][24].
- Hosting: ISO 27001 / SOC 2-certified cloud infrastructure in India, with encryption in transit and at rest [25].
5. Water Resources
India-WRIS, run by the Central Water Commission with ISRO's NRSC, is a single-window WebGIS with 12 information systems and 95 layers covering watersheds, rivers, dams, canals and reservoir levels [35]. The JRC Global Surface Water dataset, built from three million Landsat images, maps where water has appeared and disappeared since 1984 [36][37].
Water bodies are mapped with indices such as NDWI, which highlights open water but can overestimate it in built-up areas [21]; SAR is used when cloud hides the surface [15].
Business problem
- Water resources and irrigation departments need current reservoir, tank and wetland extents, but data is spread across many layers and portals [35], and checking hundreds of water bodies in the field is slow.
- Urban utilities, water-dependent industries and agri-businesses need to understand long-term water availability around their operations, and simple index maps can mislead in towns [21].
Our solution
- NDWI and SAR-based water-body mapping, with reservoir, tank and wetland trends over time.
- Long-term baselines using the JRC Global Surface Water record since 1984 [36][37], combined with official India-WRIS layers [35].
- Built-up area masking to reduce false water detection [21].
- Command-area and catchment views on WebGIS, with field checks through the My GLOBEIR app.
Benefits
- Faster status checks across many water bodies than field visits alone.
- Decades-long context: globally, permanent water disappeared from almost 90,000 sq km while 184,000 sq km of new permanent water formed between 1984 and 2015 [36].
- Fewer errors in urban areas through masking and validation [21].
- One dashboard instead of many portals [35].
Privacy & data security
- Data involved: water-body maps are non-personal. Data becomes personal when linked to farmers or landholders in an irrigation command area [30], and dam and canal layouts need careful access control.
- Access control: role-based access with separate field, supervisor and head-office roles, and audit logs.
- Stored in India: survey or drone data finer than the DST threshold stays in India and never reaches non-Indian servers [23][24]; government departments can use a MeitY-empanelled cloud, where all processing is within India [32].
- Retention and deletion: data is returned and deleted at the end of the engagement or on request.
6. Floods and Early Warning
NRSC maps flood inundation mainly from SAR because flood-season optical imagery is usually cloudy, and shares maps through Bhuvan and the National Database for Emergency Management [22]. Google's AI-based Flood Hub covered 100 countries and about 700 million people as of November 2024 [38].
SAR images day or night and through cloud [15][18]. In the Kerala 2018 study, an automatic threshold applied to Sentinel-1 images separated flood water from land [39].
Business problem
- Districts and disaster management authorities need flood-extent maps within days, but data sits across separate national products [22] and must be turned into village-level action.
- Insurers and lenders need to know which insured properties, loans or branches lie in flood-prone areas, and the RBI's draft climate-disclosure framework asks banks and large NBFCs to assess physical climate risk and geographic exposure [40].
- Utilities and businesses need to plan around repeated flooding at plants, depots and roads.
Our solution
- SAR-based flood-inundation mapping from Sentinel-1 and EOS-04 [15][18], with historical flood-frequency layers.
- Flood extent overlaid on villages, assets, insured properties or loan portfolios, shown as heatmaps and district views in administration monitoring.
- Field verification of affected areas with the My GLOBEIR app.
Benefits
- Proven accuracy: Sentinel-1 SAR flood maps of the 2018 Kerala floods reached about 94% overall accuracy (industry example) [39].
- Faster response: during Brazil's 2024 floods, Flood Hub forecasts helped aid groups distribute assistance within two days (industry example) [38].
- Longer warning: 7-day flood forecasts as accurate as earlier 5-day ones [38].
- Quantified flood exposure by asset, village or portfolio, ready for risk and disclosure reporting [40].
Privacy & data security
- Data involved: flood-extent maps are non-personal. Insured property locations, borrower addresses, relief beneficiary lists and field photos of affected homes are personal or confidential data [30].
- Minimised data: exposure analysis runs on masked, pseudonymised or aggregated data, such as village or branch totals.
- Regulated clients: for lenders, contracts can cover RBI outsourcing terms such as data storage in India and prompt incident reporting [49]; for insurers, vendor contracts and NDAs and 180-day log retention in India follow IRDAI guidelines [41].
- Stored in India: any drone or survey data finer than the DST threshold is stored and processed only in India [23][24].
7. Drought and Agriculture
The Mahalanobis National Crop Forecast Centre runs FASAL, giving in-season crop production forecasts from satellite-derived acreage and yield models, and a national agricultural drought assessment [42]. Under PMFBY crop insurance, YES-TECH gives remote-sensing yield estimates a fixed 30% weightage in claims [43].
Crop monitoring follows vegetation indices such as NDVI through the season [20], and coarse sensors that cover huge areas often suit seasonal crop monitoring.
Business problem
- Crop insurers and agriculture departments: with a fixed 30% weightage for remote-sensing yield estimates under PMFBY [43], crop estimates must be defensible.
- Agri-businesses and agri lenders need early warning of crop stress and drought in their sourcing or lending areas, not after harvest.
- Districts need drought indicators at village or block level to plan relief.
Our solution
- NDVI crop-condition time series and drought indicators by village, block or district.
- Crop-area estimation from Sentinel-2 and LISS imagery, checked against field samples [14][17].
- Geotagged field observations and crop photos with the My GLOBEIR app.
- Sourcing-area and portfolio views on WebGIS and heatmaps.
Benefits
- Early warning of crop stress during the season.
- Methods consistent with the national FASAL approach [42].
- Transparency in claims, with a fixed 30% satellite-based share in crop-insurance yield assessment [43].
- Better targeting of field visits and procurement planning.
Privacy & data security
- Data involved: crop maps are non-personal, but farmer names, plot boundaries and geotagged photos of farmers or their homes are personal data under the DPDP Act, and consent must be specific and limited to the purpose [30].
- Minimised data: crop-condition and drought analysis runs on plot IDs or aggregated village totals; farmer identifiers are not needed for most of it.
- Field app controls: background location in the My GLOBEIR app is optional and opt-in with a persistent notification; offline data syncs when connectivity returns.
- Stored in India: drone plot surveys finer than the DST threshold are stored and processed only in India [23][24].
8. Climate Change and Glaciers
ISRO's satellite study covering 1984–2023 found that 676 of 2,431 Himalayan glacial lakes larger than 10 ha had expanded notably, 130 of them in India, and 601 had more than doubled in size [44]. This is vital input for glacial-lake outburst flood risk monitoring.
Glacial lakes are mapped by comparing water extent on imagery from different years; SAR fills gaps when mountain cloud hides the surface [15].
Business problem
- Hydropower and infrastructure developers in the Himalaya have assets downstream of growing glacial lakes [44], and lenders and insurers ask about this risk.
- State disaster management authorities must watch many remote lakes that are hard to reach on the ground.
Our solution
- Multi-date glacial-lake mapping from Landsat and Sentinel-2, with SAR where cloud blocks optical imagery [13][14][15].
- Lake-growth trends and alerts for significant change.
- Downstream exposure overlays of projects, roads and villages on WebGIS.
Benefits
- Early identification of fast-growing lakes; for example, Ghepang Ghat lake in Himachal Pradesh grew 178% between 1989 and 2022 [44].
- Long baselines from the Landsat record [13].
- Remote monitoring without repeated high-altitude field trips.
- Clear exposure evidence for project risk reviews.
Privacy & data security
- Data involved: glacier and lake imagery is non-personal. Project layouts and downstream asset lists are commercially confidential.
- Stored in India: survey or drone data of project sites finer than the DST threshold is stored and processed only in India [23][24].
- Contracts: GLOBEIR signs NDAs and follows the client's information-security policies.
- Encryption: data is encrypted in transit (TLS 1.2 or higher) and at rest (AES-256).
9. Coasts and Mangroves
ISFR 2023 puts India's mangrove cover at 4,991.68 sq km [26]. Global Mangrove Watch v4.1 provides annual maps from 1985 to 2025 using optical and L-band SAR data, with a reported F1 accuracy of 0.93 [45].
Combining optical and L-band SAR data helps because optical sensors cannot see through cloud, while SAR can [15][45].
Business problem
- Forest departments and blue-carbon or restoration project developers must show where mangroves have been planted, survived or been lost.
- Port, coastal infrastructure and industrial developers need coastal clearances through PARIVESH [46], and late discovery of mangrove or coastal-zone overlaps causes delay.
- CSR teams funding restoration need independent evidence of results.
Our solution
- Mangrove extent and change mapping from Sentinel-2 and SAR, using Global Mangrove Watch as a baseline [45].
- Coastal constraint mapping around proposed sites before clearance.
- Field checks of restoration plots with the My GLOBEIR app, and progress dashboards on WebGIS.
Benefits
- Proven accuracy: global mangrove maps report an F1 of about 0.93 (industry example) [45].
- Long annual record from 1985 for baselines [45].
- Results that can be compared with national mangrove figures [26].
- Fewer surprises at clearance stage for coastal projects.
Privacy & data security
- Data involved: mangrove maps are non-personal. Field photos may show people or homes along the coast, which are personal data [30], and project boundaries before clearance are commercially confidential.
- Minimised data: field photos are captured for work purposes, and identifiable details are not needed in the reports.
- Stored in India: drone surveys of coastal sites finer than the DST threshold are stored and processed only in India [23][24].
- Access control: role-based access with audit logs of access and changes.
10. Environmental Clearance
The Environment Ministry's PARIVESH single-window portal requires project boundaries and alignments to be submitted as spatial (KML) data for environment, forest, wildlife and coastal clearances [46], enabling GIS screening against sensitive areas.
Screening works by overlaying the proposed boundary or alignment on layers of forests, protected areas, water bodies and coastal zones, and measuring overlaps and distances.
Business problem
- Project developers in infrastructure, mining, industry and real estate must submit accurate spatial data [46]. Overlaps with forests, protected areas, water bodies or coastal zones that are found late in the clearance process cause costly delays and redesign.
- Environmental consultants prepare many submissions and need consistent, correct boundary files.
Our solution
- Site-suitability and constraint mapping against forests, protected areas, water bodies and coastal zones.
- Alternative alignment comparison before design is fixed.
- KML and shapefile preparation for PARIVESH, plus drone or survey support for site boundaries.
- Post-clearance monitoring of project sites with geotagged field checks and WebGIS dashboards.
Benefits
- Fewer surprises at clearance stage, because sensitive-area overlaps are found before submission.
- Correctly formatted spatial files that match portal requirements [46].
- Faster answers to clearance queries using prepared spatial evidence.
- The same baseline layers reused for compliance monitoring after approval.
Privacy & data security
- Data involved: project boundaries and alignments before clearance are commercially sensitive, and land parcels linked to owner names are personal data [30].
- Stored in India: drone and survey data of project sites finer than the DST threshold (1 m horizontal, 3 m vertical) is stored and processed only in India and never reaches non-Indian servers [23][24].
- Contracts and audits: GLOBEIR signs NDAs and supports the client's security audits and vendor assessments.
- Deployment choice: India-hosted cloud, the client's own cloud or data centre, or on-premise.
11. Carbon and ESG
ISFR 2023 estimates India's forest carbon stock at 7,285.5 million tonnes, up 81.5 million tonnes from the previous assessment [26]. Satellite forest-change data is the standard input for monitoring deforestation-free supply chains and land-based carbon projects [29].
Satellite monitoring compares forest and land cover inside a project or sourcing area over time [29], and national estimates combine satellite mapping with field inventory [26].
Business problem
- ESG and sustainability teams must back afforestation, carbon and deforestation-free claims with independent, repeatable evidence [26][29].
- Lenders: the RBI's draft climate-disclosure framework asks banks and large NBFCs to assess physical climate risk and geographic exposure [40], but most portfolios have never been mapped against flood, cyclone or drought hazard.
- Carbon project developers need monitoring records that hold up over many years.
Our solution
- Periodic satellite monitoring of plantations, carbon projects and supply-chain sourcing areas, with geotagged field checks through the My GLOBEIR app.
- Deforestation screening of supplier locations using forest-change data [29].
- Flood, cyclone, drought and heat layers overlaid on assets or loan portfolios, shown as heatmaps.
- Exposure tables and WebGIS dashboards ready for ESG and disclosure reporting.
Benefits
- Transparent, auditable evidence for green-cover and carbon claims, using the same methods that underpin national forest accounting [26][29].
- Quantified physical-risk exposure, by asset, village or district, ready for ESG and climate-disclosure reporting [40].
- Repeatable monitoring cycles across many sites.
- Early warning of forest loss in sourcing areas.
Privacy & data security
- Data involved: forest-change imagery is non-personal. Loan portfolios, borrower locations and supplier farm locations are personal or confidential data [30].
- Minimised data: portfolio exposure analysis runs on masked, pseudonymised or aggregated data, such as district or branch totals.
- Regulated lenders: contracts can cover RBI outsourcing terms, including data storage in India, the RBI's right to inspect, provider liability and prompt incident reporting [49].
- Stored in India: project drone or survey data finer than the DST threshold is stored and processed only in India [23][24].
12. Proven Examples
- Kerala 2018 floods (peer-reviewed). Sentinel-1 SAR processed on Google Earth Engine with an automatic threshold produced flood maps with 94.3% and 94.1% overall accuracy. About 109,148 ha was under water on 9 August 2018, against 80,517–88,540 ha in August of 2015–2017 [39].
- Global surface water change (Nature, 2016). Between 1984 and 2015, permanent water disappeared from almost 90,000 sq km while 184,000 sq km of new permanent water formed elsewhere [36].
- Himalayan glacial lakes (ISRO). Ghepang Ghat lake in Himachal Pradesh grew 178%, from 36.49 ha to 101.30 ha, between 1989 and 2022 [44].
- India's national forest accounting (ISFR 2023). Net gain of 1,445 sq km of forest and tree cover since 2021, and an 81.5 Mt increase in carbon stock [26].
- Google Flood Hub. During Brazil's 2024 floods, forecasts helped aid groups distribute emergency assistance within two days [38].
13. Measurable Benefits
- Accuracy: about 94% overall accuracy for SAR flood maps [39]; F1 of about 0.93 for global mangrove maps [45].
- Lead time: 7-day flood forecasts as accurate as earlier 5-day ones [38].
- Coverage: daily global air-quality observation [16]; annual national land-use maps since 2004-05 [31]; 95 water layers in one portal [35].
- Weather independence: SAR images day or night through cloud [15][18].
- Transparency in claims: a fixed 30% satellite-based share in crop-insurance yield assessment [43].