DrillHoles

Build Status Coverage

Functions to desurvey and composite drill hole tables from the mining industry.

Installation

Get the latest stable release with Julia's package manager:

] add DrillHoles

Usage

1. Desurveying

Given a collar table, a survey table and at least one interval table (such as assay and lithology), the function desurvey can be used for desurveying. Examples of these tables are shown bellow:

tables

  • Collar table: stores the coordinates (X, Y, Z) of each drill hole with given ID (HOLEID).
  • Survey table: stores the arc length (AT) and azimuth (AZM) and dip (DIP) angles along the drill hole trajectory. Together with the collar table it fully specifies the trajectory.
  • Interval table: stores the actual measurements taken on cylinders of rock defined by an interval of arc lenghts (FROM and TO). Usually, there are multiple interval tables with different types of measurements.

Assuming that each of these tables was loaded into a Tables.jl table (e.g. CSV.File, DataFrame), we can use the following constructors to automatically detect the columns:

using DrillHoles
using CSV

collar = Collar(CSV.File("collar.csv"))
survey = Survey(CSV.File("survey.csv"))
assay  = Interval(CSV.File("assay.csv"))
litho  = Interval(CSV.File("litho.csv"))

If the columns of the tables follow an exotic naming convention, users can manually specify the names with keyword arguments:

# manually specify column with hole ID
Collar(CSV.File("collar.csv"), holeid = :MYID)

The desurvey function returns a DataFrame with standardized columns. It supports different dip angle conventions from open pit and underground mining as well as different stepping methods. Please check the docstring for more details.

df = desurvey(collar, survey, [assay, litho])

2. Compositing

The result of the desurvey function can be passed to the composite function for compositing the intervals to a given length. Please check the docstrings for available methods:

# request intervals of length 10.0
dc = composite(df, 10.0)

3. Georeferencing

The DataFrame objects produced by both desurvey and composite can be georeferenced and used as input for geostatistical modeling with GeoStats.jl:

using GeoStats

# georeference table with coordinates
georef(dc, (:X, :Y, :Z))