Validating fracture trace data
Prerequisites
Fracture or lineament trace data is typically digitized in a GIS environment. The basemap on which fractures are digitized could be an orthomosaic or a Light Detection and Ranging digital elevation model for lineaments. If topology of the fracture network is to be analyzed the constraints it poses on digitization must be taken into account. E.g. traces must be accurately snapped to end to other traces to form a Y-node (See Sanderson and Nixon, 2015).
Alongside trace data a defined target/sample area (or areas) must be supplied.
All spatial file types containing two-dimensional polylines (i.e.
LineStrings
) that can be loaded with
geopandas can be validated as trace data.
The validation tool along with all other fractopo modules only accept
geopandas.GeoDataFrame's
as inputs (geopandas
easily handles
transformation of spatial data types – shapefiles, geopackages, etc. –
to GeoDataFrames
and back). A GeoDataFrame
is the inmemory
representation of the spatial data that is modified in a Python session.
Validation
Validation consists of finding errors in digitization and then fixing
them. However currently only very few error types are automatically
fixed. Instead it is recommended to use the validation tool to find the
errors and then fixing them manually. The tool creates a new column,
VALIDATION ERRORS, in the GeoDataFrame
(visible in the attribute
table in GIS-software). Currently very few types of errors can be
automatically fixed and e.g. conversion from LineString
to
MultiLineString
has to be done to allow further validation.
Therefore I currently recommend allowing automatic fixes when prompted.
Page links below explain how to use the validation tool in Python, the validation error types and how manually fix the validation errors.
Usage examples in two notebooks:
Loading data for analysis with
geopandas
Trace validation
Visualizing found validation errors with
geopandas
andmatplotlib
For validation error descriptions: