OR/14/066 Appendix 2

Arkley, S L B, and Callaghan, E A. 2014. Model metadata report for the Culloden-Nairn GSI3D superficial deposits model. British Geological Survey Internal Report, OR/14/066.

14th March 2013 — Merging of the Culloden-Nairn and Forres models

Opened ‘Culloden_Nairn_SA_202.gsipr’ and saved as ‘Culloden_Nairn_SA_203_Forres.gsipr’.

Imported all sections from the most recent/complete Forres model, located at:
W:\Teams\QES\QMMP\Data\MorayNessBasin_Data\Data\MorayNess-Region\GSi3D\Moray Ness 3Dmodel\Published model Feb 2012\Forres_all_Hydro_Chronostrat_Eng_19Dec2011.GSIPR
using Log file located at:
W:\Teams\QES\QMMP\Data\MorayNessBasin_Data\Data\MorayNess-Region\GSi3D\Moray Ness 3Dmodel\Published model Feb 2012\Forres_Eileen_bhs_and_field_notes.blg

Imported all envelopes from the same Forres model (location described above).

Imported the DTM (.grd file) used in the Forres model, located at:
W:\Teams\QES\QMMP\Data\MorayNessBasin_Data\Data\MorayNess-Region\GSi3D\ForessLocalModelBaselineData\From Data Portal (10thJuly'09)\Forres_NEXTMAP_DTM_25m.asc
and converted it into a TIN, named ‘DTM_NextMap_25m_Forres’ and deleted the .grd file.

All geological units greyed out when completing the above actions. Looked at ‘file’ — ‘properties’ and noted that the GVS and GLEG files had converted to those of the Forres model, changed them back to the CN files (most up-to-date gvs and gleg for the Inverness area, which covers units used on both Forres and CN models) and all coloured up ok again.

SLBA identified a number of issues in merging the two models (units not matching across the boundary) — to discuss with CAA.

Meeting between SLBA and CAA to resolve merging issues — 14th March 2013

General Issues:

• Issue: Large amounts of envelope overlap along the merged edge.

• Fix: Open most recent versions of the individual models and clip the envelopes to just beyond the model edge before importing the envelopes into a ‘merged’ model, this will ensure the envelopes match with the sections of the respective models.

• Issue: Individual DTM’s were used to construct each model. The CN model used a BaldEarth 50 m resolution DTM and the F model used a NextMap 25 m resolution DTM. To calculate a model, a DTM is chosen to act as a capping surface, so currently the whole ‘merged’ area cannot be calculated as a single model.

• Fix: Need to extract a new single DTM which will cover both the CN and F models. A decision will need to be made whether NextMap or BaldEarth is used and whether a 25 m or 50 m or lower resolution is used. Immediate thoughts are that a DTM at 25 m resolution over that size of area would be too big in terms of memory required and shouldn’t even be attempted.

• Issue: A number of geological units are in different stratigraphic positions or have different names in the most up-to-date version of the GVS (used in CN model) than in the GVS used for the F model.

• Fix: Go through the two GVS’s, identify any problems with specific units (see ‘Specific Issues’ below)

• Issue: Geological unit names are not always official lexicon names, Glasgow superficial geological units were converted into lex-rcs. Note that there aren’t lexicon codes for many of the units modelled

Specific Issues:

• Issue: Sea has been modelled in CN but not in F.

• Fix: Extend existing ‘sea’ envelope from CN model across F model and draw ‘sea’ linework in all necessary F sections. Note that the ‘merged’ model will start to struggle for memory when importing topo maps from F to extend the ‘sea’ envelope — may be better to make a copy of the most recent F model, import ‘sea’ envelope from CN and extend the ‘sea’ envelope in there and then import back into the ‘merged’ model.

• Issue: Subtidal deposits have been modelled in CN but not in F.

• Fix: Extend existing ‘subtd’ envelope from CN model across F model and draw ‘subtd’ linework in all necessary F sections. Note that ‘sea’ envelope and ‘subtd’ envelope have exactly the same extent, so can probably complete the ‘sea’ envelope and then ‘copy’ (export unit as a shapefile then re-import shapefile) and paste in as a ‘subtd’ envelope.

• Issue: A number of units in the F model have a different name in the CN model, even though they are the same geological unit (e.g. hmgd=hmgd1, gfdd=gfdd1, peat=peat1, rtd6=gftd — careful with this one, I think rtd6 on F represents ‘gfsu’ glaciofluvial subglacial deposits which is different to ‘gftd’ glaciofluvial terrace deposits)

• Fix: Need to rename the units, so they have the same name/code throughout, best to change the names in the F model, as the CN model GVS is the most up-to-date. Need to make sure the envelopes for these units come across into the ‘merged’ model. See if there is a ‘find and replace’ way to change the linework codes globally, I think there is, ask Ricky, otherwise will have to go through all the sections, doing each line individually.

• Issue: A number of units modelled in F, were not in the CN model.

• Fix: Need to make sure all the envelopes come across into the ‘merged’ model.

• Issue: ‘bsa’ is in a different stratigraphic position in the F model to shown in the CN model.

• Fix: Change the GVS so there are two ‘blown sand’ units, called ‘bsa’ and ‘bsa1’, keeping the stratigraphic positions of each, i.e. 21 and 36 respectively. This should avoid any alteration of linework in the sections and is geologically realistic.

• Issue: Regional till unit in the CN model, which is the most extensive till and rests dominantly on bedrock is coded as ‘fint’, the equivalent unit in the F model is coded as ‘till6’. In the GVS these units are separated by ‘gsg6’ which is used fairly extensively in both models.

• Fix: Change ‘till6’ in the F model to ‘fint’, probably is the case geologically. See if there is a ‘find and replace’ way to change the linework codes globally, I think there is, ask Ricky, otherwise will have to go through all the sections, doing each line individually. Changes will need to be made to the nature of the till/gsg contact in F sections where the two units are on top of each other, as previously the till was the younger unit and now gsg will be the younger unit. Will also need to merge ‘till6’ envelopes and ‘fint’ envelopes, there is a function in GSI3D to do this (right click on ‘fint’ geological unit — ‘create, edit, merge’ — merge (dissolve) another unit into his unit).

• Issue: Bulk of the sand and gravel deposits at surface on the F model are coded as ‘gsg2’ and ‘gsg3’ and lie on the higher ground, away from the coast. The equivalent units in the CN model are coded as ‘gsg6’. In the GVS ‘gsg3’ and ‘gsg6’ are separated by many units, including ‘ards’ which is extensive across both models.

• Fix: Merge together ‘gsg2’ and ‘gsg3’ in the F model and rename as ‘gsg6’. Then any existing ‘gsg6’ (sand and gravels lying below ‘till6’/’fint’) change to a new unit ‘gsg7’ — will need to add this to the GVS. See if there is a ‘find and replace’ way to change the linework codes globally, I think there is, ask Ricky, otherwise will have to go through all the sections, doing each line individually. There is a function in GSI3D to merge the envelopes from two geological units, but I suspect linework will have to be deleted section by section.

• Issue: Many of the sand and gravel units appear the same colour.

• Fix: Change the sand and gravel units to different shades of pink in the GVS.

Sarah Arkley 15th March 2013
N.B. Further discussions on 28th March between Clive and Sarah, at the onset of the Beauly-Inverness modelling, may result in some changes to those written above, see notes on setting up the BI model.