MultiModel Meteogram Improvements

Our most attentive users have certainly already noticed that we re-designed our MultiModel meteogram.

The next meteogram update is live: Our MultiModel meteogram now has a new design, and also features some changes.

Here is an overview of the latest improvements:

Thinner lines for temperature and wind speed to better distinguish the individual forecasts of the different weather models. You can also select the models of interest and analyse them one by one using the model selection under the meteogram.

Smaller and thinner weather icons: The third part of the meteogram shows the weather conditions as pictograms, which we made more readable.

Model agreement analysis for precipitation: The precipitation chart was changed by showing the precipitation frequency forecast by the different forecast models. The darker the bar, the more models predict the indicated precipitation sum (shown on the y-axis). The light blue precipitation bars (red highlight in the image) show precipitation sums only predicted by one model. On the other hand, the dark blue precipitation bars (highlighted with the blue box in the image) show that 90% of all models predict a precipitation volume of 0.2mm for the indicated time frame.

We recommend using the model selection to see the precipitation sums of single models. You can select up to 5 models and see the sums individually in the chart for each model. If you test different models repeatedly in your location, you may be able to find which model is the most accurate: this is only possible by comparing each location separately, because precipitation can vary substantially within few kilometers distance.

We hope these improvements help in better understanding of the MultiModel, as well as weather developments in the most general sense, and appreciate any feedback.

Comments

Posted on 2024-04-09 12:37:59 by Benjamin.E.C.Koltenbah@#@<=%+;

I recommend a cloud cover plot showing predicted percentage cloud cover. This would best be a line plot filled down to the axis. It would provide a quick graphical view of the predicted cloud cover from the ensemble average data. Thank you for your consideration!

Posted on 2024-04-12 21:44:44 by slq5007@*+>@@+!?.com

very good!!!!!

Posted on 2024-04-14 11:50:50 by matthugo81

Improvements, but also some bad decisions - Starting with the temperature section and while the graph itself is good, the temperature scale does not change as it did previously, despite varied temperatures and forecasts. I am often left with a 3°C interval on the Y-axis and this makes reading it across as to what the likely temperatures will be difficult. Previously, the increments were far more frequent.

The ability to not be able to see the individual models for precipitation is the key change that is just not good at all. The reason is when viewing the data you can sometimes get a model 'spike' in terms of a predicted rainfall rate, which then skews the Y-axis. Because you can't individually see which model is an outlier, you now have to go through the whole individual model list to find which one is the outlier then deselect it to then alter the Y-axis to get a better view of expected precipitation.

Write comment

You need a meteoblue account to comment on articles
Back to top