LIGTAS - Learning Institution Geohazard Tracking and Assessment for Safety
Home
DepEd DRRMS × ECAIR

About LIGTAS

Learning Institution Geohazard Tracking and Assessment for Safety

A multi-hazard risk platform for Philippine schools — monitoring 48,000+ public schools across 9 hazard types with AI weather forecasting and geospatial analytics, to strengthen disaster preparedness and keep learning going through climate-related disruptions.

Explore the hazard map
Back to home
Overview

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.

What it covers

System Capabilities

48,000+ Philippine public schools monitored across 9 hazard types.

Earthquake
View map
  • PHIVOLCS real-time alerts
  • How often past shaking crossed each level (Poisson exceedance model)
  • 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 · GraphCast

24–120h forecasts with conformal uncertainty quantiles

TemperaturePrecipitationHeat IndexWindPressureHumidity
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

48,000+ public schools

Municipality and barangay resolution nationwide

Disaster Risk Reduction and Management Service (DRRMS)
Partner

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.

Learn more:

DO 50, s. 2011

|

DO 37, s. 2015
Education Center for AI Research (ECAIR)
Partner

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.

Learn more:

DepEd Order No. 13, s. 2025
Methodology & references

Methodology: An empirical exceedance-count Poisson model applied to the historical earthquake archives scraped from PHIVOLCS (2014-present). For each administrative region (municipality/barangay), LIGTAS counts how many weeks the modelled ground motion crossed each threshold, divides by the number of years observed, and turns that rate into an annual probability and a return period.

Ground motion: per-event shaking is estimated with the PHIVOLCS REDAS method (Grutas et al., 2026) and reported on the PHIVOLCS Earthquake Intensity Scale (PEIS) as the default. The legacy Boore-Joyner-Fumal (1997) attenuation with Wald et al. (1999) intensity relations is retained as a selectable comparison.

What this is not: this is not a full Probabilistic Seismic Hazard Analysis (PSHA). A PSHA requires a seismic source model and magnitude-frequency recurrence integrated over all sources, magnitudes and distances. LIGTAS does not build one. These figures describe what the recorded catalog actually did over roughly a decade, so they carry the limits of a short record and should not be used as design ground motions.

Annual Probability of Exceedance (APE)

Probability that ground motion will exceed a given threshold in one year, based on Poisson distribution model

APE=1−e−λ\text{APE} = 1 - e^{-\lambda}
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. (cited for the Poisson exceedance framing only; the source-model integration of that paper is not implemented here)

λ is estimated directly from the catalog as (number of weeks exceeding the threshold) / (years observed) - there is no fitted magnitude-frequency recurrence behind it.

Return Period

Average recurrence interval for earthquakes exceeding a given threshold, used for engineering design standards

T=1λT = \frac{1}{\lambda}
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. Intensity in the description column is reported on the PHIVOLCS Earthquake Intensity Scale (PEIS), which is the scale LIGTAS archives and reports; the Modified Mercalli Intensity (MMI) that the same PGA produces is shown in parentheses for comparison.

LevelThresholdDescription
Light
≥ 10 cm/s² (0.010 g)PEIS IV (MMI 4.5), felt by many
Moderate
≥ 50 cm/s² (0.051 g)PEIS VI (MMI 6.6), serviceability limit
Heavy
≥ 100 cm/s² (0.102 g)PEIS VII (MMI 7.5), design level
Severe
≥ 200 cm/s² (0.204 g)PEIS VIII (MMI 8.4), near-collapse
Extreme
≥ 400 cm/s² (0.408 g)PEIS VIII (MMI 9.3), very heavy damage
PGA is computed with Fukushima & Tanaka (1985)[1] as implemented in the PHIVOLCS REDAS methodology of Grutas et al. (2026)[2]; intensity uses the Richter (Gutenberg-Richter) GMICE with the MMI-to-PEIS conversion of Grutas et al. (2026)[2], Eq. 5. The legacy Boore, Joyner & Fumal (1997)[3] attenuation with the Wald et al. (1999)[4] intensity relations remains selectable for comparison.Note: PEIS and MMI are different scales and are not interchangeable. PEIS runs below MMI at high intensity (MMI 8 maps to PEIS 7.4), so an MMI value reported as a PEIS level overstates the intensity by roughly one level at the top of the scale.

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)
Exceedance Metrics (empirical Poisson)
  • 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)[8] 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[11]), 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

APE=P(X≥1 in 1 year)=1−e−λ\text{APE} = P(X \geq 1 \text{ in 1 year}) = 1 - e^{-\lambda}
where λ = exceedances / observation_years
References: Cunnane (1979)[8], Madsen et al. (1997)[9] - Standard in flood frequency analysis

Return Period

Average time interval between flood exceedances (e.g., "100-year flood" = 1% annual probability)

T=1λ (years)T = \frac{1}{\lambda} \text{ (years)}
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

ThresholdValueInfrastructure Use Case
Minor
0.10Routine maintenance planning
Moderate
0.25Drainage system design
Major
0.50Flood barrier requirements
Severe
0.75Evacuation planning
Extreme
0.90Emergency infrastructure
References: Lang et al. (1999)[10], Stedinger et al. (1993)[6], USGS Bulletin 17C (2019)[7]

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

VolcanoSummit (m asl)Location
Mayon2,463Albay, Bicol
Kanlaon2,435Negros Occidental / Oriental
Taal311 (Binintiang Malaki)Batangas, CALABARZON
Bulusan1,565Sorsogon, Bicol
Hibok-hibok1,332Camiguin
Pinatubo1,486 (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)

risk=alert_level5×proximity_factor\text{risk} = \frac{\text{alert\_level}}{5} \times \text{proximity\_factor}
proximity_factor decays exponentially beyond 3× PDZ radius
Alert LevelStatusTypical Meaning
Level 0
NormalNo unrest
Level 1
Low-Level UnrestAbnormal conditions, no eruption imminent
Level 2
Moderate UnrestMagmatic intrusion, eruption possible
Level 3
Increased UnrestEruption within weeks
Level 4
Hazardous Eruption ImminentEruption within 24h
Level 5
Hazardous Eruption OngoingLarge 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∂t+u⋅∇C=D∇2C−Cτ\frac{\partial C}{\partial t} + \mathbf{u} \cdot \nabla C = D \nabla^2 C - \frac{C}{\tau}
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

The LIGTAS AI forecasting stack runs global AI weather models daily, producing 24–120h forecasts with conformal prediction uncertainty quantiles (α = 0.05, calibrated monthly against ERA5). FourCastNetV2 is the primary model behind the displayed hazard products.

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
GraphCast  Google DeepMind
  • Graph neural network on an icosahedral mesh
  • 0.25° global forecasts, 6-hour steps
  • Autoregressive medium-range roll-out
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

CAPE=g∫zLFCzELTparcel−TenvTenv dz\text{CAPE} = g \int_{z_{\text{LFC}}}^{z_{\text{EL}}} \frac{T_{\text{parcel}} - T_{\text{env}}}{T_{\text{env}}} \, dz
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

g

Gravitational acceleration (9.81 m/s²)

Tparcel

Virtual temperature of lifted parcel

Tenv

Virtual temperature of environment

zLFC

Height of Level of Free Convection

NWS Interpretation

< 1000 J/kg

Weak instability

1000-2500 J/kg

Moderate instability

2500-4000 J/kg

Strong instability

> 4000 J/kg

Extreme instability

Lifted Index (LI)

Temperature-based instability indicator at 500 hPa following NWS classification standards

LI=Tenv(500 hPa)−Tparcel(500 hPa)\text{LI} = T_{\text{env}}(500\text{ hPa}) - T_{\text{parcel}}(500\text{ hPa})
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

> 2

Stable atmosphere

0 to 2

Marginally unstable

-2 to 0

Moderately unstable

< -6

Highly unstable

K-Index

Multi-level moisture and temperature analysis with Philippine-adjusted thresholds

K-Index=(T850−T500)+Td850−(T700−Td700)\text{K-Index} = (T_{850} - T_{500}) + T_{d_{850}} - (T_{700} - T_{d_{700}})
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

< 20

No thunderstorm expected

20-25

Isolated thunderstorms

26-30

Scattered thunderstorms

> 30

Numerous thunderstorms

Heat Index

NWS-Rothfusz equation for apparent temperature

HI=c1+c2T+c3R+c4TR+c5T2+c6R2+c7T2R+c8TR2+c9T2R2\text{HI} = c_1 + c_2 T + c_3 R + c_4 TR + c_5 T^2 + c_6 R^2 + c_7 T^2R + c_8 TR^2 + c_9 T^2R^2
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

27-32°C

Caution

33-41°C

Extreme Caution

42-51°C

Danger

> 52°C

Extreme Danger

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

Empirical exceedance-count Poisson model for PGA exceedance over the observed catalog — Cornell (1968), PHIVOLCS catalog (not a full PSHA)

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] Fukushima, Y., & Tanaka, T. (1985). A new attenuation relation for peak horizontal acceleration of strong earthquake ground motion in Japan. Bulletin of the Seismological Society of America. (PGA attenuation used by PHIVOLCS REDAS - LIGTAS default; applied as Eq. 1 of Grutas et al. 2026)

[2] Grutas, R., Dizon, R., Ramilo, R., Pabello, M. K., & Bautista, B. (2026). Evaluating the deterministic ground shaking of Camarines Norte, the Philippines, using REDAS and GIS. GeoHazards, 7(2), 41. DOI (REDAS pipeline: Eq. 1 attenuation, Eq. 3 Richter GMICE, Eq. 5 MMI-to-PEIS conversion)

[3] Boore, D. M., Joyner, W. B., & Fumal, T. E. (1997). Equations for estimating horizontal response spectra and peak acceleration from western North American earthquakes: A summary of recent work. Seismological Research Letters, 68(1), 128-153. DOI (Legacy attenuation, selectable; USGS ShakeMap OFR 2005-1135)

[4] 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 (Alternative GMICE and legacy PGA/PGV intensity relations)

Exceedance Probability (Poisson Model)

[5] Cornell, C. A. (1968). Engineering seismic risk analysis. Bulletin of the Seismological Society of America, 58(5), 1583-1606. DOI (Cited for the Poisson exceedance framing applied to earthquake and flood hazard; LIGTAS estimates the rate empirically from the observed catalog and does not implement a source-model PSHA)

Flood Hazard Analysis (Poisson Process Applications)

[6] 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)

[7] USGS. (2019). Guidelines for Determining Flood Flow Frequency. Bulletin 17C, U.S. Geological Survey Techniques and Methods 4–B5. DOI (FEMA/NFIP official methodology)

[8] 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)

[9] 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)

[10] 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)

[11] 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

Bolton, D. (1980). The Computation of Equivalent Potential Temperature. Monthly Weather Review, 108(7), 1046-1053. DOI

Rothfusz, L. P. (1990). The Heat Index Equation. NWS Technical Attachment SR 90-23. PDF

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.