Luke Skywalker and the lost Jedi Art of Regolith Mapping

Have you noticed that a number of the Star Wars stories revolve around maps? In Episode IV (the first one), R2D2 is hiding a map of the Death Star which enables the Resistance & thus Luke Skywalker to discover the secret to its "destruction".?In Episode VII, BB-8 (the millennial R2D2) is protecting the missing piece of the map that will lead to Luke Skywalker.?This missing piece allows Rey to discovery Luke on a distant planet where she’ll begin her training as a Jedi – or so she thought….

A bit of a funky segue but I see regolith mapping as the “missing piece” that will enable geologists to sample – surface & sub-surface – with a more powerful “force” that will lead to “discovery”.?And stop wasting time and money!

So, what is a regolith map??First, I think we should start with what is regolith.?I like the definition “everything between fresh rock and fresh air”.?So that encompasses soil, saprock, saprolite, alluvium, colluvium, elluvium, weathered rock (difficult for those educated on glaciated terrains I appreciate), till, lake sediment, etc……?

A regolith map to me is like a geomorphological map but simplified with an exploration “edge” added.?Excellent work by the CSIRO in Australia during the 1980s/1990s assisted in the recognition of the value of this style of mapping as a first pass exploration tool?and assisted in the discovery of multiple gold deposits in Western Australia.

The key thing for me is the boundary between transported and “in-situ”.?This is a rubbery boundary of where I’m reasonably confident surface sampling will be effective and where surface sampling may / will not work.?Note the qualifications: rubbery & reasonably – not hard and fast – I’m not going to crucify anyone that takes one or two samples of transported material.?Now the statement referencing surface sampling on transported terrain not working may incur the wrath of those who are adherents to partial leach & other techniques on transported material.?Please, put that aside for the moment – I’ll address this at the end of the post - ?as I do the “Do-Re-Mi” – let’s start at the very beginning (apologies to Julie Andrews and the cast of the Sound of Music).

So how do we start? In ye olde days (the 1990’s) we used to use stereo aerial photography, Landsat satellite data, airborne radiometrics, etc. to create a first pass regolith landform/geomorphological map.?These days with the advent of Google Maps, etc. aerial photography is less used, and I find Sentinel-2 data is my “go to” to create images for interpretation – and it’s free.?ASTER, which is my old go to, is also useful but doesn’t have the resolution of the Sentinel data.?Note Sentinel is only one of the datasets I use – aerial photography, Bing Maps, Google Earth, radiometrics, geology maps, DEM, LIDAR, shuttle radar, contours, soil maps, etc. – are also integrated.?I would love to be able to do stereo/3D interpretation but can’t quite justify the dollar outlay. Also, a new Landsat has been launched so that might be worth checking out as well.

Sentinel-2 data is available on the USGS EarthExplorer website (you need to register to download data) and reasonably simple to download. ?Landsat, ASTER and other remote sensing data are also available here.?The video below will get you started:

?Hiding in a directory ( >Granule>>Img_Data) will be the JPEG2000 images (.jp2) of the various Bands needed to create images.?The size of the images depends on their resolution – see table below - got to love 10m resolution data for free!

So now that I have my data, I bring it in to MapInfo (I believe QGIS has something similar) and create band combination images.?I prefer doing this myself rather than outsourcing because I find I need to change the colour stretch in order to optimise the regolith features I want to observe - the RASTER processor in MapInfo allows changes on the fly.?Also, I’ll start doing my interpretation as I view the images.?I’m sure MapInfo has a help video to get you started with the Raster processor so maybe have a look at that before checking out the video below:

These are the main band combinations I use to generate imagery

RGB

Natural Colour: 4 3 2

False colour Infrared: 8 4 3

Shortwave Infrared: 12 8 4

Natural Colours with Atmospheric Removal: 12 8 3

Regolith Ratio: 11/12 8a/12 8a/3

Fe: 9 8A 6

Ratios

Alteration / Laterite: 11/12

Ferric iron, Fe3+: 4/3

Ferric Oxides: 11/8

Ferrous iron: 12/8+3/4

Ferrous Silicates: 12/11

Of these I generally find the Natural Colour and the Regolith Ratio the most useful.? However, it depends upon the terrain – SWIR can be excellent in South America and Alteration/Laterite in West Africa.?Most of the time I generate all the images and find which of them are the most useful largely by trial and error.?Let me state clearly that I’m a scumbag explorationist not a specialist spectral geologist – if it works, I’ll use it; if not, it gets chucked.

OK so now you’ve got your imagery what are going to do with it??I often start by overlaying a fact geology map to identify the Basement areas.?This assists calibration of the response from the Sentinel images to what is outcrop/basement.?Drainages and areas of alluvium (generally mapped as Quaternary and/or Cainozoic) assist the identification of depositional/transported areas.?Variations in tone can have a direct correlation with Pediment or Relict/Erosional terrain.?Historic exploration, particularly drill data, should also be integrated into the interpretation at this stage.?Also, I drape my interpretation on a 3D topographic data and review it in 3D.?This is my “draft” regolith map.

Depending on the area, I’ll either use a classification of BRED or B-PIT:

·???????BRED for Basement, Relict, Erosional & Depositional

·???????B-PIT for Basement (rock sticking out of the ground), Pediment (subcrop, shallow soil cover) , Intermediate (might be P or T), & Transported.

An example below:

Then there comes field checking.?First, identify the areas of uncertainty and prioritise these on whether they are at the surface sample/do not surface sample boundary.?Second, take a shovel, auger and some sieves into the field and walk the areas.?I’m always amazed how much you “see” just by being in the field.?Walking around digging shallow holes and using sieves is an excellent way to characterise the nature of the surface material.?Also, take another geologist / field assistant (or more) in field with you to share ideas and learnings.

Then you can start planning your surface sampling program!?BRE or BPI regimes only would be where I would start and then consider aircore/RAB/auger “calibration” drilling for high priority “covered” areas.

The above is not, nor is it intended to be, a full expose on regolith mapping but hopefully it is enough to get you started and, most importantly, thinking.?

Love the rocks but don’t forget the rubble & the soil are your friends as well.

And what of the Depositional and/or Transported regime?

?If I don’t know the age, character and/or thickness of the transported material, I’m very reluctant to surface sample this regime.?If there is historic drilling where the cover sequence has been profile sampled (multi-element samples top to bottom) and it has been conclusively demonstrated that there is dispersion into the cover which has a near surface signature, then – and only then - I’ll support surface sampling.?If not, then I would consider surface sampling very risky – too many “unknown unknowns”.

It comes down to this - if you generate an anomaly in surface material in a transported regime will you be confident enough to drill it??Or does that depend on how thick the cover is? 5, 10, 50, 100 or 1000m? How do you know how thick the cover is if you haven’t drilled??A focussed program of “calibration” drilling or sampling of old drillholes will significantly increase the “knowns”.