Exploration projects collect information over the years. The information changes due to incoming data, change of personnel, among other factors. Changes in the data must be justified and well documented, it requires a sound database and professionals to provide insightful details. A simple case is to understand the impact of new logged intervals between two consecutive years using tabular data. This post refers to these tables as prior and posterior. It presents some considerations to compare to year-to-date files of categorical logs. The prior is assumed as YTD-2021 and the posterior is YTD-2022. ID is the column of drillcores ID’s. Domain is the categorical column of the interpretations. Below is an example of either the prior or posterior logging table.

Additionally, a table with the drilling year and optionally sample type is required.

The next steps prepare the files.

1. Delete leading and trailing blanks in drill core ID’s
2. Use either lower or upper case in the ID columns
3. Filter out null Domain values in the prior and posterior.

Here, the join clause uses the set of keys  ID, start and end of the interval. After the joining, the interpretation from the prior remains in the column D_prior, and in D_post for the posterior interpretations. The steps below tag the intervals with the keywords: unchanged, changed and new on a column Tag.

1. Perform a prior left outer join posterior (table 1), and tag as unchanged to equal interpretations in D_prior and D_post. This accounts for logged intervals in the prior and posterior tables that remain unchanged.

2. On table 1, tag intervals where D_prior and D_post are different, with changed. This step accounts for the intervals logged in the prior file, different in the posterior. Differences come from reinterpreting the prior (at any previous year) or deleting prior interpretations (absent in posterior).

3. Perform a prior right outer join posterior (table 2). Tag as new to null values in the column D_prior, and filter out rows where the Tag column is null. This labels interpreted intervals in the posterior file that were inexistent (null) in the prior file.  This new data comes from the same drilling year of the posterior, or from interpreting old drilling (at any previous year) logged in the year of the posterior.

4. Concatenate the rows of table 1 and 2 into a batch table with the tagging. Add a drilling year and optionally sample type column to the batch by mapping the ID’s.

The tags: unchanged, changed and new summarizes the change of information.  The length variation suffices in some cases, however is not complete. New logging data results from targeting mineralized zones, the justification of drilling and reinterpretation must be consistent to the variation of metal content. Proceed to generate a table by splitting the intervals of the logging interval table and the assays table of the elements of interest.

Apply the next steps to the new split batch table.

•  (a) Generate the statistics by D_prior.

•  (b) Generate the statistics by D_prior of the batch where Tag is changed. This is discounted from the prior due to re-interpretations.

•  (c) Get the statistics by D_post where Tag is changed, Filtering out the rows where D_post is null. This is the additioned data in the posterior that comes from reinterpreted data of the prior. It does not account for deletion of data in the posterior.

•  (d) Get the statistics of the batch where Tag is new.

•  (e) Get the statistics by domain of the entire batch (less or equal than the year of the posterior file).

The relation a - b + c + d = e holds for the length.