A new version of the DGM, REGIS II and GeoTOP models is available as of 18 March 2025. Users are advised to use these latest versions, as offered on DINOloket and BROloket (also as a download). Below we list the changes and explain them further.
In the new versions of the models (DGM v2.2.1, REGIS II v2.2.3 and GeoTOP v1.6.1), the stratigraphic and hydrogeological codes have been brought into line with those of the Stratigraphic Nomenclator. As a result, the three models now use the same coding scheme; differences have been eliminated. For GeoTOP, the new release also includes an improved estimation of the model uncertainty of the modelled geological units as well as two resolved feedbacks ('BRO Terugmeldingen’).
New encodings
Stratigraphic and hydrogeological units in the BRO models DGM, REGIS II and GeoTOP are indicated by an alphanumeric code, e.g. 'BX' for the Boxtel Formation (stratigraphic unit in DGM and GeoTOP) and 'BXk1' for the Boxtel Formation, first clay layer (hydrogeological unit in REGIS II). Formations can be further subdivided into members and beds, for example, 'NAWA' stands for Naaldwijk Formation, Walcheren Member, and 'NAWAZU' for the Zuiderzee Bed within the Walcheren Member.
What has not been reflected in the codes so far is that the formations belong to a group. This group code has now been added to the existing codes. For example, many formations in the shallow subsurface belong to the Upper North Sea Group, designated by the code 'NU'. This code 'NU' will now be added to the existing codes. The codes mentioned above will become 'NUBX', 'NUBXk1', 'NUNAWA' and 'NUNAWAZU' respectively. You can also find these codes in the Stratigraphic Nomenclature of the Netherlands. This change makes it easier to view explanatory information about a stratigraphic and hydrogeological unit in the Nomenclator.
In addition to adding the group code, a number of other changes are made. The formation code of a number of formations has changed, for example 'MT' becomes 'MA' (Maastricht Formation). Furthermore, some model unit codes in GeoTOP have been revised, for instance 'AEC' becomes 'ECga' (Echteld Formation, channel belt generation A). For more information and an overview of the changes in the codes, please refer to the explanatory memo (PDF in Dutch – 2.7MB).
Point of attention when downloading models
If you visualize the models on DINOloket, you will not notice much of the change. The content of the models remains the same as that of the previous releases. However, when you download the models and, for example, process the raster files in an automated manner, you will have to make adjustments.. This is because the code is also part of the filename of the raster files. In the SubsurfaceViewer, the codes have not yet been adjusted. This will be done at a later time.
Additional adjustments in GeoTOP
For DGM and REGIS II, the change of the coding is the only adjustment in the new release. For GeoTOP, however, the new release also contains an improved estimate of the model uncertainty of the modeled geological units. In addition, two feedbacks (‘BRO Terugmeldingen’) on GeoTOP have been resolved (see last paragraph).
The GeoTOP voxel model has two measures of model uncertainty:
- Model uncertainty of geological units – the extent to which the model is able to provide an unambiguous estimate of the geological (stratigraphic) unit to which the voxel belongs.
- Model uncertainty of lithological classes – the extent to which the model is able to provide an unambiguous estimate of the lithological class representative of the voxel.
The adjustment in GeoTOP concerns the first criterion, namely the calculation of the model uncertainty of geological units. With this adjustment the values are more realistic than before.
Model uncertainties are expressed as a number from 0 to 1, where 0 represents a very low model uncertainty (the model is well able to provide an unambiguous estimate) and 1 represents a very high model uncertainty (the model is unable to provide an unambiguous estimate). In the calculation of the model uncertainty of geological units, a scaling was applied, so that the model uncertainty in practice showed low values from 0 to about 0.5. Because these values are unrealistically low, the model uncertainty has been recalculated, omitting the scaling. The omission of the scaling leads to higher and more realistic values of the model uncertainty. In exceptional cases, the value can be higher than 1. For example, these values may occur in places where several thin layers lie on top of each other. In those places, it is very uncertain to which of these thin layers a voxel belongs.
Model uncertainty in synthetic borehole and vertical cross-section
On DINOloket you can view the new model uncertainties in the synthetic borehole and in the vertical cross-section. The new model uncertainties are also included in GeoTOP's NetCDF file. The other attributes of the voxel model (including the geological unit and the most likely lithological class) are unchanged in this release.
Please refer to the explanatory memo (PDF in Dutch – 2.7MB) for more information.
Resolved feedbacks
In the new version of GeoTOP the following feedback is resolved by including the correct grids:
- Report date 19-02-2024, description: The file of BXWISIKO_BCC (bottom of unit BXWISIKO in the layer model) is incorrect. This file shows the bottom of BXWIKO and not that of BXWISIKO.
- Report date 19-02-2024, description: In the merged layer model of GeoTOP v01r6, the grids of unit NAWO are still from an older version. These grids therefore intersect with younger or older deposits such as NAWA, NIHO and NIBA.