1/9 BLOCK MODELLING: Imagine you have a gold ore body that looks like this in section after your geologist has done his interpretation. To keep things simple, let's assume you are very promotional QP who is comfortable with just two drill holes to build your resource 😏 
2/9 In the drawing above the line representing the ore body (blue) is the geological INTERPRETATION based on known info. The block model is essentially an integration of slices (or blocks) of that interpretation so we can model all the variable properties that make up that shape.
3/9 A block model is a 3D grid of blocks in space. In our 2D example, every block that has, say >10% ore will be flagged as an ore block. Each block has the same volume, and if we know the density for each, we can calc. the tonnage inside each too.
4/9 Precision of block models is determined by size: You can see the area of blocks would be a lot bigger (more volume) if I had selected every single block that was touching the interpretation. this is the first dilemma with block models: The size of the blocks. Ideally, they..
5/9 should be as small as possible to improve precision. But this is problematic, because the number of blocks becomes so huge, it becomes difficult to manage even on super computers. The next best thing is to have the blocks size approx. the width of ore body or smallest mining
6/9 unit. Now we get to populate the blocks in the model from known information, like drilling grade info. In this case we have our composited results as shown, and they will be imprinted into the relevant blocks in 3D space.
7/9 Populating the whole model with grade in the areas represented by ore (blue) requires geostatistics. There's lies, damned lies & then there is geostatistics.😁 Key assumptions are that the ore must be homogeneous, continuous & isotropic. That's another story, and a science.
8/9 Also the sampling has to be done properly with proper QA/QC. Having done the geostats, all your blocks containing ore now have grade in them too. So with the densities and the volumes, the sumproduct of which becomes your Measured & Indicated Resource.
9/9 So each block in your model has a grade value, waste is set to zero obviously. But each block can have other attributes as well, such as density, mining sequence, geotechnical info, cost info and recovery info. They become very powerful & useful tools at any mine.
@threadreaderapp unroll
