Graphical Modelling and Causal Inference

There is a newer version of the record available.

Published May 1996 | Version v1
Dataset Open

Hailfinder

Description

This datset is set about severe summer hail in northeastern Colorado. It was introduced in Abramson et al. (1996).

 

Task: The dataset can be used to study causal discovery algorithms.

 

Summary: 

  • Size of dataset: 20,000 x 56
  • Task: Causal Discovery
  • Data Type: Categorical
  • Dataset Scope: Standalone
  • Ground Truth: Unknown
  • Temporal Structure: Static
  • License: TBD
  • Missing Values: No

 

Missingness Statement: There are no missing values.

 

Features:

  • N07muVerMo: 10.7mu vertical motion (StrongUp, WeakUp, Neutral, and Down)
  • SubjVertMo: subjective judgment of vertical motion (StrongUp, WeakUp, Neutral, and Down)
  • QGVertMotion: quasigeostrophic vertical motion (StrongUp, WeakUp, Neutral, and Down)
  • CombVerMo: combined vertical motion (StrongUp, WeakUp, Neutral, and Down)
  • AreaMesoALS: area of meso-alpha (StrongUp, WeakUp, Neutral, and Down)
  • SatContMoist: satellite contribution to moisture (VeryWet, Wet, Neutral, and Dry)
  • RaoContMoist: reading at the forecast center for moisture (VeryWet, Wet, Neutral, and Dry)
  • CombMoisture: combined moisture (VeryWet, Wet, Neutral, and Dry)
  • AreaMoDryAir: area of moisture and dry air (VeryWet, Wet, Neutral, and Dry)
  • VISCloudCov: visible cloud cover (Cloudy, PC, and Clear)
  • RCloudCover: infrared cloud cover (Cloudy, PC, and Clear)
  • CombClouds: combined cloud cover (Cloudy, PC, and Clear)
  • CldShadeOth: cloud shading, other (Cloudy, PC, and Clear)
  • AMInstabMt: AM instability in the mountains (None, Weak, and Strong)
  • InsInMt: instability in the mountains (None, Weak, and Strong)
  • WndHodograph: wind hodograph (DCVZFavor, StrongWest, Westerly, and Other)
  • OutflowFrMt: outflow from mountains (None, Weak, and Strong)
  • MorningBound: morning boundaries (None, Weak, and Strong)
  • Boundaries: boundaries (None, Weak, and Strong)
  • CldShadeConv: cloud shading, convection (None, Some, and Marked)
  • CompPlFcst: composite plains forecast (IncCapDecIns, LittleChange, and DecCapIncIns)
  • CapChange: capping change (Decreasing, LittleChange, and Increasing)
  • LoLevMoistAd: low-level moisture advection (StrongPos, WeakPos, Neutral, and Negative)
  • InsChange: instability change (Decreasing, LittleChange, and Increasing)
  • MountainFcst: mountains (region 1) forecast (XNIL, SIG, and SVR)
  • Date: date (May15_Jun14, Jun15_Jul1, Jul2_Jul15, Jul16_Aug10, Aug11_Aug20, and Aug20_Sep15)
  • Scenario: scenario (A, B, C, D, E, F, G, H, I, J, and K)
  • ScenRelAMCIN: scenario relevant to AM convective inhibition (AB and CThruK)
  • MorningCIN: morning convective inhibition (None, PartInhibit, Stifling, and TotalInhibit)
  • AMCINInScen: AM convective inhibition in scenario (LessThanAve, Average, and MoreThanAve)
  • CapInScen: capping within scenario (LessThanAve, Average, and MoreThanAve)
  • ScenRelAMIns: scenario relevant to AM instability (ABI, CDEJ, F, G, H, and K)
  • LIfr12ZDENSd: LI from 12Z DEN sounding (LIGt0, N1GtLIGt_4, N5GtLIGt_8, and LILt_8)
  • AMDewptCalPl: AM dewpoint calculations, plains (Instability, Neutral, and Stability)
  • AMInsWliScen: AM instability within scenario (LessUnstable, Average, and MoreUnstable)
  • InsSclInScen: instability scaling within scenario (LessUnstable, Average, and MoreUnstable)
  • ScenRel34: scenario relevant to regions 2/3/4 (ACEFK, B, D, GJ, and HI)
  • LatestCIN: latest convective inhibition (None, PartInhibit, Stifling, and TotalInhibit)
  • LLIW: LLIW severe weather index (Unfavorable, Weak, Moderate, and Strong)
  • CurPropConv: current propensity to convection (None, Slight, Moderate, and Strong)
  • ScnRelPlFcst: scenario relevant to plains forecast (A, B, C, D, E, F, G, H, I, J, and K)
  • PlainsFcst: plains forecast (XNIL, SIG, and SVR)
  • N34StarFcst: regions 2/3/4 forecast (XNIL, SIG, and SVR)
  • R5Fcst: region 5 forecast (XNIL, SIG, and SVR)
  • Dewpoints: dewpoints (LowEverywhere, LowAtStation, LowSHighN, LowNHighS, LowMtsHighPl, HighEverywhere, and Other)
  • LowLLapse: low-level lapse rate (CloseToDryAd, Steep, ModerateOrLe, and Stable)
  • MeanRH: mean relative humidity (VeryMoist, Average, and Dry)
  • MidLLapse: mid-level lapse rate (CloseToDryAd, Steep, and ModerateOrLe)
  • MvmtFeatures: movement of features (StrongFront, MarkedUpper, OtherRapid, and NoMajor)
  • RHRatio: relative humidity ratio (MoistMDryL, DryMMoistL, and Other)
  • SfcWndShfDis: surface wind shifts and discontinuities (DenvCyclone, E_W_N, E_W_S, MovigFtorOt, DryLine, None, and Other)
  • SynForcng: synoptic forcing (SigNegative, NegToPos, SigPositive, PosToNeg, and LittleChange)
  • TempDis: temperature discontinuities (QStationary, Moving, None, and Other)
  • WindAloft: wind aloft (LV, SWQuad, NWQuad, and AllElse)
  • WindFieldMt: wind fields, mountains (Westerly and LVorOther)
  • WindFieldPln: wind fields, plains (LV, DenvCyclone, LongAnticyc, E_NE, SEquad, and WidespdDnsl)

Files

hailfinder.csv

Files (10.7 MB)

Name Size
md5:48da72236fe15de48c7e7ddd94b5c278
10.7 MB Preview Download

Additional details

Related works

Is documented by
10.1007/978-1-4757-3502-4_6 (DOI)

References

  • Abramson B, Brown J, Edwards W, Murphy A, Winkler RL (1996). "Hailfinder: A Bayesian System for Forecasting Severe Weather." International Journal of Forecasting, 12(1):57–71.
  • Elidan, G. Bayesian Network Repository. (2001), https://www.cs.huji.ac.il/w~galel/Repository/
  • Beinlich I, Suermondt HJ, Chavez RM, Cooper GF (1989). "The ALARM Monitoring System: A Case Study with Two Probabilistic Inference Techniques for Belief Networks". Proceedings of the 2nd European Conference on Artificial Intelligence in Medicine, 247–256.
  • Scutari M (2010). "Learning Bayesian Networks with the bnlearn R Package." Journal of Statistical Software, 35(3), 1–22. doi:10.18637/jss.v035.i03