Height Data - Accuracy
The accuracy of a height dataset is usually detailed in terms of vertical accuracy. However, it is important to recognise that there is also a horizontal accuracy associated with any height dataset.
A variety of factors can effect the accuracy of a dataset depending on the chosen method of data collection.
LiDAR
Currently available LiDAR data has been collected with a stated vertical accuracy of up to +/- 0.10 metres RMSE and horizontal accuracies to within +/-0.80 metres RMSE. A vertical accuracy of +/- 0.15 metres RMSE is, at present, frequently quoted as the accepted standard.
When collecting height data using LiDAR techniques there are a number of variables which may effect accuracy levels such as the equipment used, the altitude and speed of the aircraft and the weather conditions at the time the data is collected. Furthermore the type of terrain for which the data is being collected will also influence accuracy levels. Rapid changes in terrain or steeply sloping terrain results in higher levels of inaccuracy. This, to some extent, can be mitigated by collecting the data at a higher resolution, thereby minimising the risk that a rapid change in terrain is simply missed and not recorded. Areas of continuous and dense building and tree coverage also poses problems for LiDAR data, making it particularly difficult to filter and interpret the data collected in order to generate a reliable DTM.
IFSAR
Data collected by IFSAR is generally less accurate than data collected by LiDAR. It has acheived stated vertical accuracies of up to +/- 1 metre RMSE and horizontal accuracies of up to +/- 2.5 metres RMSE.
As the collection method for IFSAR is broadly the same (using sound waves rather than light waves to measure distance), most of the same factors which effect the accuracy of LiDAR data also effect data collected by IFSAR.
Photogrammetry
Typically, height data derived from photogrmmetry is more accurate than LiDAR and IFSAR data in the x and y (horizontal) direction but less accurate in the z (vertical) direction.
Photogrammetry is frequently more reliable in areas which pose difficulties for the collection of LiDAR and IFSAR data. Factors such as steep or rapid changes in terrain and the coverage of buildings causes fewer problems to the accuracy of photogrammetric data. For instance, photos can clearly define a ridge or the edges of a building when the point cloud footprint from LiDAR and IFSAR cannot. Conversely, photogrammetry is relatively less reliable in flat and featureless areas. In consequence, some datasets have used photogrammetry in conjunction with LiDAR in order to incorporate the advantages of both systems and integrate it to create a better product.
Topographical Survey
Current GPS survey techniques provide highly accurate data, with achievable vertical accuracies of within +/- 10 mm RMSE.