About LIGTAS
Learning Institution Geohazard Tracking and Assessment for Safety
Project Overview
LIGTAS is a comprehensive geospatial analytics research initiative developed through the partnership between the Disaster Risk Reduction and Management Service (DRRMS) and the Education Center for AI Research (ECAIR), established under DepEd Order No. 13, s. 2025.
The system develops multi-hazard mapping systems for schools, addressing landslide, flood, heat, volcanic, and other environmental risks to enhance disaster preparedness and preserve learning continuity during climate-related disruptions.
System Capabilities
47,000+ Philippine public schools monitored across 9 hazard types
Earthquake
View Map →- • PHIVOLCS real-time alerts
- • Probabilistic Seismic Hazard Analysis (PSHA)
- • PGA climatology per municipality
Flood
View Map →- • Sentinel-1 SAR weekly flood maps
- • Poisson return period analysis
- • Annual probability of exceedance
Typhoon
View Map →- • Vorticity + pressure gradient detection
- • LPA formation probability
- • Real-time track monitoring
Volcanic
View Map →- • PHIVOLCS alert level risk (6 volcanoes)
- • In-browser ashfall simulation (24h)
- • Volcanic seismic PGA (Boore 2014)
Landslide
View Map →- • ERA5 rainfall-triggered susceptibility
- • Weekly Zarr archive per municipality
- • Slope + soil saturation factors
Storm Surge
View Map →- • Coastal exposure index
- • Typhoon wind-driven surge risk
- • Municipality-level classification
Heat Index
View Map →- • NWS-Rothfusz equation
- • PAGASA danger thresholds
- • AI forecast heat index (24-120h)
Tsunami
View Map →- • Elevation + coastal distance model
- • PHIVOLCS susceptibility zones
- • Municipality-level susceptibility
Drought
View Map →- • ERA5 precipitation deficit index
- • Weekly archive per municipality
- • Cumulative dry-spell tracking
Platform Capabilities
AI Weather Forecasting
FourCastNetV2 · Aurora · Pangu-Weather
24–120h forecasts with conformal uncertainty quantiles
Satellite Imagery
Himawari-9 real-time
IR / visible / water vapor channels
Atmospheric Stability
CAPE · Lifted Index · K-Index
NWS/WMO standards with PH-adjusted thresholds
School Coverage
47,000+ public schools
Municipality and barangay resolution nationwide
About DRRMS
Disaster Risk Reduction and Management Service
The Disaster Risk Reduction and Management Service (DRRMS) serves as DepEd's focal and coordinative unit for disaster risk reduction, emergency response, and climate change adaptation across all educational institutions in the Philippines.
Mission
- • Institutionalize the culture of safety at all levels
- • Systematize protection of education investments and ensure continued delivery of quality education services
- • Serve as the focal and coordinative unit for DRRM-related activities
Core Functions
- • Focal point for DRRM, Education in Emergencies (EiE), and Climate Change Adaptation (CCA)
- • Develop and recommend policy standards on DRRM/EiE/CCA matters
- • Coordinate with NGAs, NGOs, CSGs, and NDRRMC Technical Working Groups
- • Enhance DepEd's resilience to disasters through policy development
- • Lead Education Cluster and Protection Group initiatives
LIGTAS Partnership Role
- • Operational requirements & use case definition
- • Field validation & testing protocols
- • Integration with DepEd disaster response systems
- • Policy guidance & compliance standards
- • Multi-stakeholder coordination & deployment
- • Real-world emergency response applications
Mandates: Established through DepEd Order No. 50, s. 2011 (Creation of DRRMO) and DO 37, s. 2015 (Comprehensive DRRM in Basic Education Framework). DRRMS ensures LIGTAS aligns with national disaster risk reduction policies and serves the practical needs of schools nationwide.
About ECAIR
Education Center for AI Research
The Education Center for AI Research (ECAIR) was established under DepEd Order No. 13, s. 2025 to support research and development in artificial intelligence applications for Philippine basic education.
For the LIGTAS project, ECAIR handles technical implementation including system architecture, AI model integration, and geospatial data processing. This work is done in partnership with DRRMS, which provides operational guidance and ensures alignment with DepEd's disaster risk management needs.
Methodology: Standard Probabilistic Seismic Hazard Analysis (PSHA) equations using historical earthquake archives from PHIVOLCS. System computes long-term seismic hazard statistics per administrative region (municipality/barangay).
Annual Probability of Exceedance (APE)
Probability that ground motion will exceed a given threshold in one year, based on Poisson distribution model
where λ = exceedances / observation_years
Reference: Cornell, C. A. (1968). Engineering seismic risk analysis. Bulletin of the Seismological Society of America, 58(5), 1583-1606.
Seismic hazard analysis: McGuire, R. K. (2004). Seismic Hazard and Risk Analysis. Earthquake Engineering Research Institute, Monograph MNO-10, 221 pp.
Return Period
Average recurrence interval for earthquakes exceeding a given threshold, used for engineering design standards
where T = return period in years
Engineering Design Examples:
- • 475-year return period = 10% probability of exceedance in 50 years (typical building design)
- • 2,475-year return period = 2% probability in 50 years (critical facilities)
- • Used by NBCP (National Building Code of the Philippines) for seismic design
PGA (Peak Ground Acceleration) Thresholds
NBCP-based thresholds for seismic design and hazard classification
| Level | Threshold | Description |
|---|---|---|
| Light | ≥ 10 cm/s² | MMI III-IV, felt by many |
| Moderate | ≥ 50 cm/s² | MMI V-VI, serviceability limit |
| Heavy | ≥ 100 cm/s² | MMI VII-VIII, design level |
| Severe | ≥ 200 cm/s² | MMI IX+, near-collapse |
| Extreme | ≥ 400 cm/s² | MMI X+, very heavy damage |
References: Abrahamson & Silva (2008)[1], Wald et al. (1999)[2]
Climatology Output Statistics
Per administrative region (municipality/barangay), computed from historical earthquake archives
PGA Statistics
- • Mean, median, maximum, std. dev
- • Percentiles: 10th, 25th, 50th, 75th, 90th, 95th, 99th
- • Standard statistical distribution (no arbitrary weights)
PSHA Metrics
- • APE for all 5 PGA thresholds
- • Return periods (years) for each threshold
- • Annual event rates
- • Total events observed
MMI (Modified Mercalli Intensity)
- • Mean and maximum MMI values
- • Correlated from PGA using Wald et al. (1999)
- • Describes shaking intensity and damage potential
Data Sources
- • PHIVOLCS earthquake catalog
- • Weekly Zarr archives aggregated monthly
- • CF-1.8 compliant metadata
Methodology: LIGTAS applies Poisson distribution for probabilistic flood hazard assessment. The Poisson process has been the standard framework in hydrology since Cunnane (1979)[6] for modeling flood occurrence in Partial Duration Series (PDS) and Peaks-Over-Threshold (POT) analysis[7][8]. This approach is widely used for infrastructure design (Habeeb & Bastidas-Arteaga, 2023[9]), return period estimation, and risk-based decision making in urban drainage and floodplain management.
Annual Probability of Exceedance (APE)
Probability that flood threshold will be exceeded at least once per year based on Poisson distribution
where λ = exceedances / observation_years
References: Cunnane (1979)[6], Madsen et al. (1997)[7] - Standard in flood frequency analysis
Return Period
Average time interval between flood exceedances (e.g., "100-year flood" = 1% annual probability)
Infrastructure Design Examples:
- • 10-year flood = 10% annual probability (routine drainage design)
- • 50-year flood = 2% annual probability (major infrastructure)
- • 100-year flood = 1% annual probability (FEMA flood insurance standard)
Flood Probability Thresholds
Five severity levels for probabilistic flood hazard assessment
| Threshold | Value | Infrastructure Use Case |
|---|---|---|
| Minor | 0.10 | Routine maintenance planning |
| Moderate | 0.25 | Drainage system design |
| Major | 0.50 | Flood barrier requirements |
| Severe | 0.75 | Evacuation planning |
| Extreme | 0.90 | Emergency infrastructure |
References: Lang et al. (1999)[8], Stedinger et al. (1993)[4], USGS Bulletin 17C (2019)[5]
Practical Applications
Engineering-grade flood risk metrics for infrastructure planning and budget allocation
Infrastructure Design
Determine flood barrier height using return periods. 20-year protection requires barriers above severe threshold.
Budget Allocation
Prioritize funding for barangays with high APE. Target areas with >50% annual major flood probability.
Emergency Response
Activate pre-emptive evacuation when current forecast exceeds extreme threshold with high APE.
Data Source
S1Flood weekly archives aggregated monthly. Cloud-optimized Zarr format with CF-1.8 metadata.
Methodology: Three complementary models — PHIVOLCS alert-level risk (operational), volcanic seismic PGA climatology (Boore 2014 GMPE with low-Q crustal correction), and an in-browser ashfall finite-difference simulation for the 6 WOVOdat-monitored volcanoes.
Monitored Volcanoes
Six PHIVOLCS / WOVOdat-monitored volcanoes with real-time bulletin scraping every 30 minutes
| Volcano | Summit (m asl) | Location |
|---|---|---|
| Mayon | 2,463 | Albay, Bicol |
| Kanlaon | 2,435 | Negros Occidental / Oriental |
| Taal | 311 (Binintiang Malaki) | Batangas, CALABARZON |
| Bulusan | 1,565 | Sorsogon, Bicol |
| Hibok-hibok | 1,332 | Camiguin |
| Pinatubo | 1,486 (post-1991 caldera rim) | Zambales / Pampanga / Tarlac |
Alert Level Risk Model
Municipality risk derived from PHIVOLCS alert levels (0–5) and proximity to the Permanent Danger Zone (PDZ)
proximity_factor decays exponentially beyond 3× PDZ radius
| Alert Level | Status | Typical Meaning |
|---|---|---|
| Level 0 | Normal | No unrest |
| Level 1 | Low-Level Unrest | Abnormal conditions, no eruption imminent |
| Level 2 | Moderate Unrest | Magmatic intrusion, eruption possible |
| Level 3 | Increased Unrest | Eruption within weeks |
| Level 4 | Hazardous Eruption Imminent | Eruption within 24h |
| Level 5 | Hazardous Eruption Ongoing | Large eruption in progress |
Ashfall Simulation
In-browser finite-difference advection-diffusion solver estimating ashfall extent from current PHIVOLCS eruption column height and AI-forecast surface winds
C = ash concentration · u = wind vector · D = diffusivity · τ = settling time
Model Parameters
- • Simulation window: 6–24h (capped at 24h)
- • Time step: 6h frames
- • Wind input: u10/v10 from active AI forecast model
- • Source height: from PHIVOLCS bulletin or alert-level default
Plume Height Defaults
- • Alert Level 1: 500 m above vent
- • Alert Level 2: 1,000 m above vent
- • Alert Level 3: 3,000 m above vent
- • Alert Level 4–5: 5,000–8,000 m above vent
Note: Column height is reported above the crater rim (above vent), not above sea level. Heights sourced from PHIVOLCS bulletins parsed in real time; summit elevations from PHIVOLCS / NAMRIA.
AI Forecast Models
Three state-of-the-art AI weather models run daily, producing 24–120h forecasts with conformal prediction uncertainty quantiles (α = 0.05, calibrated monthly against ERA5)
FourCastNetV2
NVIDIA / ECMWF
- • Spherical Fourier Neural Operator
- • Primary model — flood risk probability, humidity
- • 73 atmospheric variables at 0.25° resolution
Aurora
Microsoft Research
- • 3D Swin Transformer architecture
- • High-resolution precipitation and wind
- • Pretrained on ERA5 + CMIP6 + GFS
Pangu-Weather
Huawei Cloud
- • 3D Earth Transformer
- • Hierarchical temporal aggregation (1/3/6/24h)
- • Optimized for tropical cyclone track
Data source: ERA5 reanalysis (ECMWF) used as initial conditions. Forecasts archived as Zarr (CF-1.8) in Google Cloud Storage and served directly to the browser via chunked HTTP range requests.
CAPE (Convective Available Potential Energy)
Measures atmospheric instability and thunderstorm potential using 13-level vertical integration from surface to 50 hPa
Reference: Moncrieff, M. W., & Miller, M. J. (1976). The dynamics and simulation of tropical cumulonimbus and squall lines. Quarterly Journal of the Royal Meteorological Society, 102(432), 373-394. DOI: 10.1002/qj.49710243208
Virtual temperature correction: Doswell, C. A., & Rasmussen, E. N. (1994). The effect of neglecting the virtual temperature correction on CAPE calculations. Weather and Forecasting, 9(4), 625-629. DOI: 10.1175/1520-0434(1994)009<0625:TEONTV>2.0.CO;2
NWS Interpretation:
Lifted Index (LI)
Temperature-based instability indicator at 500 hPa following NWS classification standards
Reference: Galway, J. G. (1956). The Lifted Index as a predictor of latent instability. Bulletin of the American Meteorological Society, 37(10), 528-529.
NWS operational use: Johns, R. H., & Doswell, C. A. (1992). Severe local storms forecasting. Weather and Forecasting, 7(4), 588-612. DOI: 10.1175/1520-0434(1992)007<0588:SLSF>2.0.CO;2
NWS Classification:
K-Index
Multi-level moisture and temperature analysis with Philippine-adjusted thresholds
Reference: George, J. J. (1960). Weather Forecasting for Aeronautics. Academic Press. (Original K-Index formulation)
Severe weather application: Brooks, H. E., Lee, J. W., & Craven, J. P. (2003). The spatial distribution of severe thunderstorm and tornado environments from global reanalysis data. Atmospheric Research, 67, 73-94. DOI: 10.1016/S0169-8095(03)00045-0
Philippine-Adjusted Thresholds:
Heat Index
NWS-Rothfusz equation for apparent temperature
Reference: Rothfusz, L. P. (1990). The Heat Index Equation (or, More Than You Ever Wanted to Know About Heat Index). NWS Technical Attachment SR 90-23. National Weather Service, Southern Region Headquarters, Fort Worth, TX.
PAGASA implementation: PAGASA. (2022). Heat Index Categories and Health Advisories. Philippine Atmospheric, Geophysical and Astronomical Services Administration. pagasa.dost.gov.ph
PAGASA Heat Index Categories:
Output Variables
18 total variables including 13 web-displayable with physics-constrained superresolution (4× enhanced spatial detail)
Typhoon Detection
- • Typhoon Formation Probability
- • Wind Speed Analysis
- • LPA Formation Metrics
- • Vorticity Analysis
Thunderstorm Analysis
- • CAPE Calculation
- • Lifted Index
- • K-Index
- • Thunderstorm Probability
Heat & Precipitation
- • Heat Index (NWS-Rothfusz)
- • Total Precipitation
- • Flood Risk Analysis
- • Relative Humidity
NWS/WMO-Compliant Analysis
Our system implements National Weather Service (NWS) and World Meteorological Organization (WMO) standard methodologies for atmospheric stability assessment, with thresholds adjusted for Philippine tropical conditions in consultation with PAGASA.
- • CAPE: Convective energy measurement using 13-level vertical integration
- • Lifted Index: Temperature-based instability indicator at 500 hPa
- • K-Index: Multi-level moisture and temperature analysis
Hazard-Specific Algorithms
Typhoon Detection
Vorticity + pressure gradient + wind speed analysis from AI forecast models
Thunderstorm Analysis
CAPE + Lifted Index + K-Index composite scoring (NWS/WMO standards, PH-adjusted)
Heat Index
Rothfusz (1990) equation with PAGASA danger thresholds applied to AI forecast output
Flood Hazard Climatology
Poisson Partial Duration Series (PDS) model for return period estimation — Cunnane (1979), Madsen et al. (1997)
Earthquake Hazard Climatology
Probabilistic Seismic Hazard Analysis (PSHA) with Poisson process for PGA exceedance — Cornell (1968), PHIVOLCS catalog
Volcanic Hazard
PHIVOLCS alert-level risk × proximity decay + in-browser ashfall FD simulation (24h, 6h steps) + Boore (2014) low-Q seismic PGA
Landslide Susceptibility
ERA5 cumulative rainfall trigger model with slope and soil saturation factors, weekly Zarr archive
Seismology & Earthquake Hazard
- [1] Abrahamson, N. A., & Silva, W. J. (2008). Summary of the Abrahamson & Silva NGA ground-motion relations. Earthquake Spectra, 24(1), 67-97. DOI
- [2] Wald, D. J., et al. (1999). Relationships between peak ground acceleration, peak ground velocity, and modified Mercalli intensity in California. Earthquake Spectra, 15(3), 557-564. DOI
Probabilistic Hazard Analysis (PSHA)
- [3] Cornell, C. A. (1968). Engineering seismic risk analysis. Bulletin of the Seismological Society of America, 58(5), 1583-1606. DOI (Applied to both earthquake and flood hazard assessment)
- McGuire, R. K. (2004). Seismic Hazard and Risk Analysis. Earthquake Engineering Research Institute, Monograph MNO-10, 240 pp.
Flood Hazard Analysis (Poisson Process Applications)
- [4] Stedinger, J. R., Vogel, R. M., & Foufoula-Georgiou, E. (1993). Frequency analysis of extreme events. Handbook of Hydrology, McGraw-Hill, Chapter 18. (Standard reference for flood frequency analysis)
- [5] USGS. (2019). Guidelines for Determining Flood Flow Frequency. Bulletin 17C, U.S. Geological Survey Techniques and Methods 4–B5. DOI (FEMA/NFIP official methodology)
- [6] Cunnane, C. (1979). A note on the Poisson assumption in partial duration series models. Water Resources Research, 15(2), 489-494. DOI (Foundational work on Poisson distribution for flood occurrence)
- [7] Madsen, H., Rasmussen, P. F., & Rosbjerg, D. (1997). Comparison of annual maximum series and partial duration series methods for modeling extreme hydrologic events: 1. At-site modeling. Water Resources Research, 33(4), 747-757. DOI (Comparative analysis of PDS Poisson models)
- [8] Lang, M., Ouarda, T. B. M. J., & Bobée, B. (1999). Towards operational guidelines for over-threshold modeling. Journal of Hydrology, 225(3-4), 103-117. DOI (Operational POT guidelines with Poisson validation)
- [9] Habeeb, B., & Bastidas-Arteaga, E. (2023). Assessment of the impact of climate change and flooding on bridges and surrounding area. Frontiers in Built Environment, 9. DOI (Infrastructure design with stochastic Poisson process)
Atmospheric Stability Indices
- Moncrieff, M. W., & Miller, M. J. (1976). The dynamics and simulation of tropical cumulonimbus and squall lines. Quarterly Journal of the Royal Meteorological Society, 102(432), 373-394. DOI
- Doswell, C. A., & Rasmussen, E. N. (1994). The effect of neglecting the virtual temperature correction on CAPE calculations. Weather and Forecasting, 9(4), 625-629. DOI
- Galway, J. J. (1956). The Lifted Index as a predictor of latent instability. Bulletin of the American Meteorological Society, 37(10), 528-529. DOI
- George, J. J. (1960). Weather Forecasting for Aeronautics. Academic Press, 673 pp.
Thermodynamics & Heat Index
Severe Weather Forecasting
- Johns, R. H., & Doswell, C. A. (1992). Severe local storms forecasting. Weather and Forecasting, 7(4), 588-612. DOI
- Brooks, H. E., Lee, J. W., & Craven, J. P. (2003). The spatial distribution of severe thunderstorm and tornado environments from global reanalysis data. Atmospheric Research, 67-68, 73-94. DOI
Philippine Standards
- PAGASA. (2022). Heat Index Categories and Health Advisories. Philippine Atmospheric, Geophysical and Astronomical Services Administration. Link
- PAGASA. (2022). Tropical Cyclone Wind Signal. Link
- National Building Code of the Philippines (NBCP). (2015). National Structural Code of the Philippines (NSCP) Volume I. 7th Edition.
- PHIVOLCS. (2024). Earthquake Information. Department of Science and Technology. Link
All citations follow American Meteorological Society (AMS) style guidelines. Complete references available in REFERENCES.md.
© 2025 ECAIR × DRRMS. Learning Institution Geohazard Tracking and Assessment for Safety.