"Because its a good general value to describe the height of the building. The value of the 70th percentile tends to be closer to the ridge of the roof than the min. or the 50th percentile, and excludes the height of the chimneys or other objects on the roof. This it what most people are interested in I think, so we display the h_dak_70p value in the viewer. And including all the h_dak percentiles would be too much info in the viewer. Note, that the data downloads contain all the h_dak values.
Additionally, we also use the 70th percentile for the data sets that we are developing for noise simulations (3DGeluid), because we need a roof height value that approaches the real ridge of the roof as close as possible, but excludes the extra object on top of the roof.
But keep in mind that the h_dak values are "just" percentiles of the roof height, and not an exact reference to particular parts of the roof (eaves, ridge etc.).
All this said, all the other height percentiles, min, 50, max are just as good and valid choice as the 70th. It all depends on what do you need the height values for. For instance, if you are interested in the lowest height of the roof (eg. for taxation, energy use estimation etc), then the min is a good choice. If you want to know the elevation at which is safe for drones to fly around, then you need the max height. As you see, it all depends on your use case.
"
And how does that relate to the data in general.
"Because its a good general value to describe the height of the building. The value of the 70th percentile tends to be closer to the ridge of the roof than the min. or the 50th percentile, and excludes the height of the chimneys or other objects on the roof. This it what most people are interested in I think, so we display the h_dak_70p value in the viewer. And including all the h_dak percentiles would be too much info in the viewer. Note, that the data downloads contain all the h_dak values. Additionally, we also use the 70th percentile for the data sets that we are developing for noise simulations (3DGeluid), because we need a roof height value that approaches the real ridge of the roof as close as possible, but excludes the extra object on top of the roof.
But keep in mind that the h_dak values are "just" percentiles of the roof height, and not an exact reference to particular parts of the roof (eaves, ridge etc.).
All this said, all the other height percentiles, min, 50, max are just as good and valid choice as the 70th. It all depends on what do you need the height values for. For instance, if you are interested in the lowest height of the roof (eg. for taxation, energy use estimation etc), then the min is a good choice. If you want to know the elevation at which is safe for drones to fly around, then you need the max height. As you see, it all depends on your use case. "