# Why are tornadoes so difficult to predict?

###### November 19, 2015 3:56PM

We do not know everything about the weather. We do however know how most weather-formations build up. This is also true for the conditions causing tornadoes. The problem involves accurate predictions as in the actual weather forecasts.

We can predict tornadoes by watching the skies. This can be done by satellite's, by radar, from ground, or a combination. This is all adding accuracy to the predictions.

The biggest problems of accurate weather forecasts is the amount of data needed to be calculated in order to provide predictions. We know to a certain extent how to use the data in calculations but this is still an inaccurate science at best. Some even believe that a butterfly flapping its wings at the wrong place at the right time can influence the weather on the opposite side of our globe.

The smallest changes in weather one place can change the outcome another place. We live on a living planet and as of that we simply have to accept certain conditions.

As we make weather models to support the globe as it is today, these models may have to be changed in the future as we change our globe due to deforestation and building dams, cities etc.

The most advanced models of our world today are able to predict weather up to 30 days forward. Not to a great accuracy as for which specific day and time it will start to rain, but within a few days. This sort of predictions is used in Shipping and Oil industry.

As of being able to actually calculate (I would rather use that word than "predict", because it all boils down to calculations). We can not "predict" a certain weather or a certain condition only for a small geographic area. This add uncertainty to predictions. The tornado touching down may be 3-10 miles away from where it was thought to be.

The amount of data needed for exact predictions/calculations are simply not available or possible to collect. To illustrate this further, consider the network of measuring-stations on ground. We can improve this network by placing one station for every 100 Square meters. This could be a kind of "back bone" in a new statistical model. Even so, with such a network, we would not be able to calculate proper predictions.

Reasons: The lower atmosphere tell one story, but is dependent upon higher altitude atmospheric conditions. Let's say we one day manage to measure (I don't know how, just lets say it is possible) the atmospheric conditions for every 100 meter up into the outer edge of our atmosphere.If we had all this statistical information readily available, then all we need to do is to put this into a statistical model, and we would have a really great tool.But even so!We need to be able to calculate accurately, and we can add much more information into this statistical model.We need to know the position of the moon and its distance from earth at all times. We need to know the position of all the other planets in order to calculate gravitational fields. Ultimately, we need to use all this information in combination with data from the sun. Is there going to be a solar flare? will the solar flare be directed so that it will be caught up in gravitational fields and be led towards earth in part or in full.It may sound like science fiction, but everything is calculateable if we have the power to do it and the knowledge as in how to do it.The amount of data needed for accurate calculations is immense. I do however beleive that it can be done.

One method could be to force an issue to the producers of mobile phones and computer equipment. Governments could force measurement-capability into all units, and the units could send data of measurement every time the measurements change. This would be able to provide very localized data for most of the globe where people are. It would still not be enough though.