Temporal, probabilistic mapping of ash clouds using wind field stochastic variability and uncertain eruption source parameters: Example of the 14 <scp>A</scp>pril 2010 <scp>E</scp>yjafjallajökull eruption

Stefanescu, E. R.; Patra, Abani; Bursik, Marcus; Madankan, Reza; Pouget, S.; Jones, Matthew D.; Singla, Puneet; Singh, Tarunraj; Pitman, E. Bruce; Pavolonis, Michael J.; Morton, Don; Webley, P. W.; Dehn, J.

doi:10.1002/2014ms000332

Cited by 26 publications

(21 citation statements)

References 40 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The difficulties in observing volcanic eruptions and constraining the properties of volcanic ash clouds means that there is still significant uncertainty associated with the ESPs used to initialize operational dispersion models, which propogates into the volcanic ash advisories and ash concentration charts issued. Future operational systems should consider implementing probabilistic ash cloud forecasting, which explores the variability in the ESPs, for example by using an ensemble approach [132,133].…”

Section: Modelling Volcanic Ash In the Atmospherementioning

confidence: 99%

Atmospheric Dispersion Modelling at the London VAAC: A Review of Developments since the 2010 Eyjafjallajökull Volcano Ash Cloud

et al. 2020

View full text Add to dashboard Cite

It has been 10 years since the ash cloud from the eruption of Eyjafjallajökull caused unprecedented disruption to air traffic across Europe. During this event, the London Volcanic Ash Advisory Centre (VAAC) provided advice and guidance on the expected location of volcanic ash in the atmosphere using observations and the atmospheric dispersion model NAME (Numerical Atmospheric-Dispersion Modelling Environment). Rapid changes in regulatory response and procedures during the eruption introduced the requirement to also provide forecasts of ash concentrations, representing a step-change in the level of interrogation of the dispersion model output. Although disruptive, the longevity of the event afforded the scientific community the opportunity to observe and extensively study the transport and dispersion of a volcanic ash cloud. We present the development of the NAME atmospheric dispersion model and modifications to its application in the London VAAC forecasting system since 2010, based on the lessons learned. Our ability to represent both the vertical and horizontal transport of ash in the atmosphere and its removal have been improved through the introduction of new schemes to represent the sedimentation and wet deposition of volcanic ash, and updated schemes to represent deep moist atmospheric convection and parametrizations for plume spread due to unresolved mesoscale motions. A good simulation of the transport and dispersion of a volcanic ash cloud requires an accurate representation of the source and we have introduced more sophisticated approaches to representing the eruption source parameters, and their uncertainties, used to initialize NAME. Finally, upper air wind field data used by the dispersion model is now more accurate than it was in 2010. These developments have resulted in a more robust modelling system at the London VAAC, ready to provide forecasts and guidance during the next volcanic ash event.

show abstract

Section: Modelling Volcanic Ash In the Atmospherementioning

confidence: 99%

Atmospheric Dispersion Modelling at the London VAAC: A Review of Developments since the 2010 Eyjafjallajökull Volcano Ash Cloud

et al. 2020

View full text Add to dashboard Cite

show abstract

“…Earth scientists also employ the ROC curve for a diverse set of modeling activities, including the distribution of rock glaciers (e.g., Brenning et al, 2007), assessing triggering mechanisms of earthquake aftershocks (e.g., Meade et al, 2017), and snow slab instability physics (e.g., Reuter & Schweizer, 2018). This also includes land‐air interactions, such as mapping of expected ash cloud locations after eruptions (e.g., Stefanescu et al, 2014), modeling rainfall‐induced landslides (e.g., Anagnostopoulos et al, 2015), and statistically forecasting extreme corn losses in the eastern United States (Mathieu & Aires, 2018). The fields of space and planetary science have also started to employ this technique, such as for oblique ionogram retrieval algorithm assessment (Ippolito et al, 2016), identifying energetic particle flux injections at Saturn (e.g., Azari et al, 2018), magnetic activity prediction (e.g., Liemohn, McCollough, et al, 2018), and identifying solar flare precursors (e.g., Chen et al, 2019).…”

Section: Introductionmentioning

confidence: 99%

The STONE Curve: A ROC‐Derived Model Performance Assessment Tool

Liemohn

Azari

Ganushkina

et al. 2020

Earth and Space Science

View full text Add to dashboard Cite

A new model validation and performance assessment tool is introduced, the sliding threshold of observation for numeric evaluation (STONE) curve. It is based on the relative operating characteristic (ROC) curve technique, but instead of sorting all observations in a categorical classification, the STONE tool uses the continuous nature of the observations. Rather than defining events in the observations and then sliding the threshold only in the classifier/model data set, the threshold is changed simultaneously for both the observational and model values, with the same threshold value for both data and model. This is only possible if the observations are continuous and the model output is in the same units and scale as the observations, that is, the model is trying to exactly reproduce the data. The STONE curve has several similarities with the ROC curve—plotting probability of detection against probability of false detection, ranging from the (1,1) corner for low thresholds to the (0,0) corner for high thresholds, and values above the zero‐intercept unity‐slope line indicating better than random predictive ability. The main difference is that the STONE curve can be nonmonotonic, doubling back in both the x and y directions. These ripples reveal asymmetries in the data‐model value pairs. This new technique is applied to modeling output of a common geomagnetic activity index as well as energetic electron fluxes in the Earth's inner magnetosphere. It is not limited to space physics applications but can be used for any scientific or engineering field where numerical models are used to reproduce observations.

show abstract

“…Earth scientists also employ the ROC curve for a diverse set of modeling activities, including the distribution of rock glaciers (e.g., Brenning et al, 2007), assessing triggering mechanisms of earthquake aftershocks (e.g., Meade et al, 2017), and snow slab instability physics (e.g., Reuter & Schweizer, 2018). This also includes land-air interactions, such as mapping of expected ash cloud locations after eruptions (e.g., Stefanescu et al, 2014), modeling rainfall-induced landslides (e.g., Anagnostopoulos et al, 2015), and statistically forecasting extreme corn losses in the eastern United States (Mathieu & Aires, 2018). The fields of space and planetary science have also started to employ this technique, such as for oblique ionogram retieval algorithm assessment (Ippolito et al, 2016), identifying energetic particle flux injections at Saturn (e.g., Azari et al, 2018), magnetic activity prediction (e.g., Liemohn, McCollough, et al, 2018), and identifying solar flare precursors (e.g., Chen et al, 2019).…”

Section: Introductionmentioning

confidence: 99%

The STONE curve: A ROC-derived model performance assessment tool

Liemohn

Azari²,

Ganushkina³

et al. 2020

Preprint

View full text Add to dashboard Cite

Key Points:• A new event-detection-based metric for model performance appraisal is given with sliding thresholds in both observational and model values• The new metric is like the relative operating characteristic curve but uses continuous observational values, not just categorical status• The new metric is used on real-time model predictions of common geomagnetic activity parameters, demonstrating its features and strengths AGU Index Terms:• 1984 Statistical methods: Descriptive (4318)• 4318 Statistical analysis (1984, 1986) • 7924 Forecasting (1922, 2722, 4315) • 0550 Model verification and validation• 9820 Techniques applicable in three or more fields Abstract A new model validation and performance assessment tool is introduced, the sliding threshold of observation for numeric evaluation (STONE) curve. It is based on the relative operating characteristic (ROC) curve technique, but instead of sorting all observations in a categorical classification, the STONE tool uses the continuous nature of the observations. Rather than defining events in the observations and then sliding the threshold only in the classifier/model data set, the threshold is changed simultaneously for both the observational and model values, with the same threshold value for both data and model. This is only possible if the observations are continuous and the model output is in the same units and scale as the observations, i.e., the model is trying to exactly reproduce the data. The STONE curve has several similarities with the ROC curve -plotting probability of detection against probability of false detection, ranging from the (1,1) corner for low thresholds to the (0,0) corner for high thresholds, and values above the zero-intercept unity-slope line indicating better than random predictive ability. The main difference is that the STONE curve can be nonmonotonic, doubling back in both the x and y directions. These ripples reveal asymmetries in the data-model value pairs. This new technique is applied to modeling output of a common geomagnetic activity index as well as energetic electron fluxes in the Earth's inner magnetosphere. It is not limited to space physics applications but can be used for any scientific or engineering field where numerical models are used to reproduce observations. Plain Language SummaryScientists often try to reproduce observations with a model, helping them explain the observations by adjusting known and controllable features within the model. They then use a large variety of metrics for assessing the ability of a model to reproduce the observations. One such metric is called the relative operating characteristic (ROC) curve, a tool that assesses a model's ability to predict events within the data. The ROC curve is made by sliding the eventdefinition threshold in the model output, calculating certain metrics and making a graph of the results. Here, a new model assessment tool is introduced, called the sliding threshold of observation for numeric evaluation (STONE) curve. The STONE curve is created by sliding t...

show abstract

Temporal, probabilistic mapping of ash clouds using wind field stochastic variability and uncertain eruption source parameters: Example of the 14 April 2010 Eyjafjallajökull eruption

Cited by 26 publications

References 40 publications

Atmospheric Dispersion Modelling at the London VAAC: A Review of Developments since the 2010 Eyjafjallajökull Volcano Ash Cloud

Atmospheric Dispersion Modelling at the London VAAC: A Review of Developments since the 2010 Eyjafjallajökull Volcano Ash Cloud

The STONE Curve: A ROC‐Derived Model Performance Assessment Tool

The STONE curve: A ROC-derived model performance assessment tool

Contact Info

Product

Resources

About