Compositing Drill Intervals

Downhole samples usually arrive at irregular lengths, which makes grades hard to compare and biases anything that treats every sample equally. Compositing rebuilds the data as longer intervals whose values are the length-weighted average of the samples they cover, written as a downhole table of their own. Two methods are offered: Fixed interval composites to a regular length, and Significant intersection finds the mineralised intercepts along the existing sample boundaries. The same dialog can also derive a metal equivalent - a single grade combining several assays using commodity prices and recoveries.

To open it, click the data processing button at the right of the Drill Manager toolbar and choose Composite.... The dialog works on that Drill Manager's DrillDB file.

Source

Choose the downhole interval Dataset to composite and tick the columns to carry into the composites. Only numeric columns can be composited; text columns such as lithology are not offered.

Fixed interval

Choose Fixed interval as the Method.

• Interval length - the regular downhole length of each composite, measured from the start depth. Boundaries fall at start, start + length, start + 2 x length and so on.
• Start depth - where the first boundary is registered. Usually 0.
• Minimum residual length - the leftover interval at the end of a hole's data is kept only when it is at least this long; a shorter tail is dropped.
• Minimum coverage % - two rules share this value. A composite interval is only written when at least this fraction of it is covered by source samples, so a window falling in a gap in the data is skipped rather than written empty. And within a written interval, a column's value is only written when at least this fraction is covered by samples that are not null - otherwise the cell is null.

Each written value is the length-weighted mean of the samples overlapping the interval, weighted by the overlapping length only, with null samples contributing neither value nor weight.

Significant intersection

Choose Significant intersection as the Method to report grade intercepts the way they are quoted in drilling announcements - "12.0 m @ 2.10 g/t Au incl. 3.0 m @ 6.50 g/t Au". Intercepts follow the existing sample boundaries; the samples are not split.

• Grade column - the grade the intercepts are selected on. When a metal equivalent is enabled below, its output column is offered too and is calculated per sample before compositing.
• Cut-off grade - the minimum length-weighted grade of an intercept, with its internal dilution included.
• Dilution grade - samples below this are internal dilution. An intercept always starts and ends on a sample at or above it. While Dilution grade same as cut-off is ticked it equals the cut-off; untick it and set a lower value to let weakly mineralised samples join an intercept, as long as the whole still makes the cut-off.
• Max consecutive dilution - the longest continuous length of dilution an intercept may include. A longer run of low grade ends the intercept.
• Max total dilution - the most dilution one intercept may include in all. 0 for no limit.
• Minimum length - shorter intercepts are not reported.
• Minimum grade x length - intercepts whose grade times length (gram-metres, percent-metres) is below this are not reported. 0 for no limit.
• Top cut - sample grades above this are capped to it before compositing, so a single very high assay cannot carry an intercept. 0 for uncut.
• Missing samples - unsampled gaps and null grades inside an intercept are either treated as Zero-grade dilution, counted against the dilution limits, or Break the intercept so that no intercept crosses them.
• Including cut-off - when set, the same rules are applied again inside each intercept at this higher cut-off, and the higher-grade parts are written to the Including table and added to the intercept's label. 0 for none.

Intercepts are found from the top of the hole down. Each one starts at the shallowest qualifying sample, then is extended across dilution to the deepest qualifying sample that keeps its grade at the cut-off - a weak stretch followed by strong grade is still taken in when the whole makes the cut-off.

Each intercept row carries the grade column as tested (top-cut and diluted), the other ticked columns as their length-weighted mean over non-null samples, and four summary columns: Length, the grade x length, Dilution (the internal dilution and missing length) and Intercept, the text label.

Output

The composites are written to a downhole interval table of their own: under Output table, give a new table a name or choose an existing one to replace. When an including cut-off is set, also name the Including table. Replacing rewrites the whole table - composite boundaries move whenever the parameters change, so the old rows cannot be kept.

Metal equivalent

A metal equivalent expresses several assays as one grade: each commodity's contribution is its grade times its price and recovery, and the total is divided by the value of one grade unit of the base commodity. Copper equivalent from copper and gold, for example, reads in the copper row's grade unit.

Tick Calculate a metal equivalent, then add a row per commodity with + Commodity: its name, the assay column it comes from, the unit that column is in, the price and its unit, and the metallurgical recovery in percent. Grade units convert as 1% = 10000 ppm, with g/t and ppm equal; prices per troy ounce and per pound are converted to price per tonne of contained metal.

Choose the Base commodity the equivalent is expressed in, and name the Output column.

The table of prices can be saved under a name with Save As... and chosen again later from Price table. Price tables are kept in the .si/commodityPrices folder of your home directory, so they are shared by all your projects. The column mappings are saved with the table; a mapping that names a column the current file lacks shows as unmapped.

Choose where the equivalent is written: Write into the composite table calculates it from the composited grades, and Write onto the source table calculates it for each original sample. You can tick both. If any contributing assay is null the equivalent is null, unless Treat missing grades as zero is ticked.