The model output animation and accompanying forecast table is generated using the HYSPLIT numerical dispersion model. The model uses estimates of volcano emissions along with forecast winds to predict the concentrations of sulfur dioxide gas (SO2) and sulfate aerosol particles (SO4) downwind of the ongoing Kilauea eruption. This is a research effort that is in progress. Disclaimer.
The probability of exceeding 0.01 ppm SO2 or 1 ug/m3 SO4 (and other thresholds like 0.1 and 1 ppm SO2; 15 and 65 ug/m3 SO4) plots show how likely a "trace" of vog is to be detected in the area and hour shown, and likewise for the other thresholds.
The ensemble mean plots show the average conditions after running the vog model 27 times with each run having slightly different conditions than the others.
The ensemble variance plots show much the 27 vog model runs differ from each another. A large variance also shows a lot of uncertainty in the forecast.
The regular/deterministic forecast shows the forecast from the original single vog model run.
|Thu, Nov 14, 2019||1300||0.00||0.00||0.00||0.00||0.00||0.00||0.00||0.00||0.00||0.00||0.00|
|Thu, Nov 14, 2019||1600||0.00||0.00||0.00||0.00||0.00||0.00||0.00||0.00||0.00||0.00||0.00|
|Thu, Nov 14, 2019||1900||0.00||0.00||0.00||0.00||0.00||0.00||0.00||0.00||0.00||0.00||0.00|
|Thu, Nov 14, 2019||2200||0.00||0.00||0.00||0.00||0.00||0.00||0.00||0.00||0.00||0.00||0.00|
|Fri, Nov 15, 2019||0100||0.00||0.00||0.00||0.00||0.00||0.00||0.00||0.00||0.00||0.00||0.00|
|Fri, Nov 15, 2019||0400||0.00||0.00||0.00||0.00||0.00||0.00||0.00||0.00||0.00||0.00||0.00|
|Fri, Nov 15, 2019||0700||0.00||0.00||0.00||0.00||0.00||0.00||0.00||0.00||0.00||0.00||0.00|
|Fri, Nov 15, 2019||1000||0.00||0.00||0.02||0.00||0.00||0.00||0.00||0.00||0.00||0.00||0.00|
|Fri, Nov 15, 2019||1300||0.00||0.00||0.02||0.00||0.00||0.00||0.00||0.00||0.00||0.00||0.00|
|Fri, Nov 15, 2019||1600||0.00||0.00||0.01||0.00||0.00||0.00||0.00||0.00||0.00||0.00||0.00|
|Fri, Nov 15, 2019||1900||0.00||0.00||0.00||0.00||0.00||0.00||0.00||0.00||0.00||0.00||0.00|
|Fri, Nov 15, 2019||2200||0.00||0.00||0.01||0.00||0.00||0.00||0.00||0.00||0.00||0.00||0.00|
|Sat, Nov 16, 2019||0100||0.00||0.00||0.00||0.00||0.00||0.00||0.00||0.00||0.00||0.00||0.00|
|Thu, Nov 14, 2019||1300||0||0||0||0||0||0||0||0||0||0||0|
|Thu, Nov 14, 2019||1600||0||0||0||0||0||0||0||0||0||0||0|
|Thu, Nov 14, 2019||1900||0||0||0||0||0||0||0||0||0||0||0|
|Thu, Nov 14, 2019||2200||0||0||0||0||0||0||0||0||0||0||0|
|Fri, Nov 15, 2019||0100||0||0||0||0||0||0||0||0||0||0||0|
|Fri, Nov 15, 2019||0400||0||0||0||0||0||0||0||0||0||0||0|
|Fri, Nov 15, 2019||0700||0||0||0||0||0||0||0||0||0||3||0|
|Fri, Nov 15, 2019||1000||0||0||0||0||0||0||0||0||0||0||0|
|Fri, Nov 15, 2019||1300||0||0||0||0||0||0||0||0||0||0||0|
|Fri, Nov 15, 2019||1600||0||0||0||0||0||0||0||0||0||0||0|
|Fri, Nov 15, 2019||1900||0||0||0||0||0||0||0||0||0||0||0|
|Fri, Nov 15, 2019||2200||0||0||0||0||0||0||0||0||0||0||0|
|Sat, Nov 16, 2019||0100||0||0||0||0||0||0||0||0||0||0||0|
The new graphics now include contours of the topography (lines that connect areas of equal elevation) to display important features of the terrain.What is the vog forecast ensemble?
An ensemble is a group of items viewed together instead of individually. In this case, the ensemble is a group of vog forecasts. We are now running the vog model 27 times! When you view the entire ensemble of forecasts together, it can create a better forecast and provide information about how 'certain' the model is. For example, imagine a lot of forecasters in a room together. When individual people produce forecasts each forecaster may have considered something a little different than the other people in the group. By combining forecasts together and averaging them, often a better forecast is created than any one forecaster working alone. Not only that, but how much the forecasts differ from each other provides information about how certain the group is of their forecast. The ensemble average (or mean) represents the average concentrations predicted by all 27 members together. The ensemble spread (or variance) provides information about where the forecasts disagree or differ from each other and by how much.What is the Probability of Exceeding (#), and how is it calculated?
The probability of exceeding the different thresholds indicate how likely the air quality will be as bad as or worse than the threshold indicated. These are calculated using the vog model forecast ensemble. The graphics below do not indicate the severity of vog but the probability of encountering vog-influenced air with concentrations that exceed each threshold. For example, in the SO2 graphics below (threshold of 0.01 ppm SO2), if red colors (>50%) are shown over the south Kona area, there is a 75-50% chance the air over the area will exceed a concentration of 0.01 ppm of SO2.Why were these thresholds selected for the probability of exceedance plots?
The lowest thresholds (0.01 ppm SO2, 1 ug/m3 SO4) shown in the graphics below were selected to indicate the presence of air that has been exposed to vog. The other thresholds (0.1 ppm SO2/15 ug/m3 SO4; 1 ppm SO2/65 ug/m3 SO4;) respectively indicate air quality that falls into the 'Moderate or Worse' and 'Unhealthy or Worse' categories.