Welcome to Defra Data Services Forum

    What is the vertical accuracy of your LIDAR data?

    Follow

    Our specifications require the absolute height error to be less than ±15cm. This is the root mean squared error or RMSE. It quantifies the error or difference between the Ground Truth Survey and the LIDAR data. We expect the relative height error (random error) to be no more than ±5cm.

    To find out the Ground Truth result of a particular survey please see our LIDAR Capture Programme dashboard.

     

     

    Was this article helpful?
    0 out of 0 found this helpful

    Comments

    Please sign in to leave a comment

    • Avatar
      Data Services Platform team

      Question:
      What is the vertical accuracy for the more historic data? From the shapefile index catalogues I have found that some of the composite DTM data I am using is from various years as far back as 2006, despite being in the 2019 and 2020 tile sets. Please could you confirm the accuracies for the data through the years?

      Answer:
      All our time series LIDAR that goes into the production of our LIDAR Composite products is compared against an independent ground control (https://experience.arcgis.com/experience/753ad2ebd3554fa696885b8c366c3049/page/page_16/?views=view_19).

      Our QC process requires the absolute height error of every LIDAR survey to be less than ±15cm against at least one independent ground control. The independent control is undertaken by a field surveyor recording elevation values using a highly accurate RTK receiver. Surveys are usually carried out on flat open areas, such as car parks, with around 200-300 elevation recordings taken at random. Each ground truth elevation recording has a vertical accuracy of +/-0.03 m. These elevations are then compared against the LIDAR survey and the root mean squared error, or RMSE calculated. The RMSE quantifies the error, or difference, between the predicted values, i.e. the Ground Truth Survey, and the observed values, the LIDAR data.

      Please note it is possible that when the composite is produced with two overlapping surveys there could be upto a 30cm difference between elevation values in the same location if one survey has a positive bias of 15cm against ground control and the second a negative bias of 15cm. Historically older surveys are also more likely to have a slightly poorer comparison against control. This is due to less accurate RTK positions and more error in the older LIDAR instruments. The average error against ground control for all our surveys over the past 5 years has been around +/-5cm RMSE.

      We hope this helps.